This project investigated the effects of two different work-rest schedules on fatigue and performance. The study addressed an important problem for Australian drivers and the long distance road transport industry in particular, that of driver fatigue. There is considerable evidence that for professional long distance drivers, fatigue is a major problem that is not being adequately addressed by current working hours regulations. Alternative approaches to fatigue management have been proposed and are in operation in Queensland, however these approaches have been hampered by a lack of specific evidence on what represents an effective work-rest schedule for managing fatigue.
In this project, so far, two work-rest schedules were chosen to examine their effects on fatigue and performance capacity. The first study looked at the influence of working the current working hours regulatory regime on fatigue and performance. This was designed to provide a baseline for comparison with alternative approaches to work-rest scheduling. The second study looked at an alternative work-rest schedule that involved drivers doing significantly longer trips than permitted under the regulatory regime. The first study was an on-road study where fatigue and performance were measured at intervals while drivers were working. In contrast, the second study was conducted as a simulation as the work-rest schedule being evaluated in this study contravened the working hours regulations. The simulation involved drivers going through all aspects of the trip, but not on the road.
The results of the first study showed that drivers working under the regulated hours regime did not report particularly high levels of fatigue over a work day nor over a work week. Over the first single shift these findings were mirrored by performance effects. There was no clear evidence of performance deterioration in this group of drivers whose trip lasted on average between 13 and 14 hours. The results suggest also that for a single trip of this length the effects on performance are the same for day and night work, provided drivers had a long rest immediately before the trip.
Over the work week, performance results were also much like the fatigue results since performance remained at much the same level as it had been at the beginning of the trip. There was, however, some performance deterioration across the week in the ability to pick up infrequent visual signals. The results of the Mackworth Vigilance test showed a significantly greater number of missed signals at the end of the week compared to the beginning. Compared to the performance standard developed against alcohol effects in the earlier laboratory studies (CR 189), this level of decrement was not sufficiently large to be of concern for safety. Nevertheless, it was concluded that this result is likely to be an early warning of more significant performance deteriorations that would occur in work-schedules with longer work periods and less rest over the work week.
Where the regulated regime study showed no changes of immediate concern, the results of the simulation study showed significant deterioration in fatigue and performance over the much longer trip. Furthermore, performance was significantly poorer than found for the laboratory-based alcohol standard for 0.05%BAC, thus leading to the conclusion that this trip would not be safe if it were conducted on the road. The long trips that were done in the simulation showed high levels of fatigue and clear effects on performance that would be very likely to affect driving, particularly affecting drivers' ability to pick up signals. The simulation trip also showed the consequence of not balancing rest with work demands in very long trips since after the second long day of work and with only around six hours overnight sleep, drivers did not recover and performance effects occurred very soon after starting the third day.
This conclusion was supported by the findings of the on-road study of consistent relationships between work, rest, fatigue and performance over a work shift and over the work week. Over a work shift, both fatigue and performance were positively related to the number of hours worked indicating that controlling the length of the work shift is appropriate for managing fatigue over a single shift. The results of this study do not allow us to be specific about the maximum number of hours that can be done with safety. The simulation study suggested that up to 16 hours of work may be done by rested drivers without producing significant adverse effects on performance, but that such a long shift cannot be sustained for longer than one shift. This finding needs to be replicated on the road as overall performance levels in this simulation were considerably poorer than expected.
From the on-road study, over a work week control of fatigue and safe performance were related to the steps drivers took to manage fatigue even at quite low levels of fatigue. Drivers who took more breaks and obtained more sleep had lower fatigue and showed better performance. The simulation revealed that recovery from fatigue is dependent on the accumulated level of fatigue and that recovery may not occur when too much fatigue had accumulated and insufficient recovery time was allowed. So it seems from these results that in the longer term, effective fatigue management should emphasise rest-taking rather than necessarily only limiting the length of the work shift.
These studies also provide a model for evaluating the effects of long hours on fatigue and performance. Using a set of performance tests that have demonstrated sensitivity for fatigue, based on laboratory results, allows interpretation of the meaning of the results for safety. Because the tests have been standardised, it is possible to conclude from the results of the working hours regime evaluation that the small performance change across the work week did not compromise safety, but from the extended trip simulation it must be concluded that performance during that trip would be highly likely to be unsafe.
A number of issues have not been resolved from these studies. While the results suggest that trips of around 14 hours and possibly even up to 16 hours do not compromise safe performance for rested drivers, it was not possible for this report to look at changes in performance within the trip. This issue needs to be addressed further. Similarly, the results of the on-road working hours regime study showed only very moderate effects on performance, but most drivers in this study did considerably shorter total weekly hours than are allowed by the working hours regulations. From these results it is not possible to suggest what the effects of longer weekly hours would be.
Clearly then, the results of these studies have shed light on the relative effects of a number of important aspects of the working hours regime on fatigue and safe performance. Not surprisingly, a number of issues need further clarification, but the results of these studies also show that we now have an effective method for evaluating the effects of work-rest scheduling on fatigue and performance.
These studies were important for three reasons. Most notably, these findings provide evidence about the effectiveness of two different work-rest regimes and suggest some general principles for the management of fatigue. The results showed that trips that are conducted under the current working hours regime of up to 14 hours allowable in each trip, do not produce significant levels of fatigue nor significant adverse effects on performance. The results of both studies show, however, that increasing hours of work produce both fatigue and poorer performance, even over the week. Further work is needed to be specific about how many hours work can be done and still remain safe. The simulation study suggests that up to 16 hours of work may be done without producing significant adverse effects on performance, but that such a long trip cannot be sustained for longer than one trip. This finding needs to be replicated on the road as overall performance levels in this simulation were considerably poorer than expected.
These studies also showed the importance of taking into account the effects of accumulation of fatigue over consecutive trips. The simulation revealed that recovery from fatigue is dependent on the accumulated level of fatigue and that recovery may not even occur when too much fatigue has accumulated. The on-road study showed that even at relatively low levels of fatigue, the amount of sleep and the number of breaks determine fatigue and performance. The results imply that management of chronic fatigue will be achieved best by controlling the amount and pattern of rest breaks rather than simply the number of working hours permitted.
The second reason that these findings are important is that the observed performance effects are of relevance to driving. Performance functions, such as reaction speed and the ability to consistently react quickly, and the ability to detect infrequent signals, were affected most by work-rest experiences and fatigue. Decrements in these types of functions are certainly likely to affect driving ability. It is also significant that the same performance effects, increases in the number of missed signals in the Mackworth Vigilance test and more variable reaction speed in the Simple Reaction Time test, were found across work periods in both studies. This suggests that these functions are most sensitive to fatigue.
Lastly, these studies are important because they demonstrate a method for systematically evaluating the effects of different work-rest patterns on fatigue and performance. One of the major problems for management of fatigue in the long distance road transport industry has been a lack of scientific evidence for recommending effective limits for work or patterning of rest. The results of these studies and of the previous laboratory study demonstrate a method that is sensitive enough to pick up effects of variations in the work-rest schedule and variations in fatigue levels. This method will therefore enable informed judgements to be made about how work and rest can be arranged to reduce the problem of driver fatigue. This method can be applied to assess other work-rest schedules, to evaluate their effectiveness for managing fatigue and to develop alternative schedules where a particular schedule is shown to be ineffective.
Type: Research and Analysis Report
Sub Type: Consultant Report
Author(s): A Williamson, A-M Feyer, R Friswell, S Finlay Brown
ISBN: 0 642 25579 2
Topics: Fatigue, Heavy vehicle
Publication Date: 01/06/00