contact us | site help | site map
levies and fees | publications | forms

Incident Reporting

AUSREP

REEFVTS

Declared Vessels

Great Barrier Reef and Torres Strait

Levies and Fees

Standards and Regulations

Navigation Safety

Vessel Tracking

OHS Inspectorate

Port State Control

Flag State Administration

Coastal Pilotage

• Great Barrier Reef pilotage fatigue risk assessment
• Fatigue study on coastal pilots
• The work practices of marine pilots
• Work schedules of Great Barrier Reef Pilots
• Impact on wives and families
• Work and Sleep patterns
• Implications for fatigue management
• Information for Interest
• Restricted access

Codes Manuals and Reports

You are here: Home >

On Tour analyses of the work and rest patterns of Great Barrier Reef pilots: implications for fatigue management

Results: PATTERNS OF ACTIVITY ON THE SHIP

6.0 Bridgework

The previous sections and earlier stages of the project have provided a description of the general characteristics of work assignments. This section focuses on the features of bridgework including the timing and duration, extent of bridgework in critical time periods and differences in these measures across the 3 shipping routes.

6.1 Total Duration, Number and Duration of each Bridge Period per Assignment

An average assignment involved bridgework which totalled 21.8 hours on the Inner Route, 7.45 hours on Hydrographers Passage and 8.9 hours on the GNE Channel (Figure 6.0). The significantly longer total duration of bridge periods on the Inner Route reflects the longer assignment duration of 54 hours for this route.

The extended hours of bridgework on the Inner Route was completed over an average of 7 periods. In contrast, on the shorter routes bridgework averaged 1 period, with this occasionally being more than one. The length of individual bridge periods averaged 3 hours on the Inner Route compared with 6.6 hours on the other routes (Table 6.0). This finding indicates that the shorter routes involved one longer period of bridgework in contrast with multiple bridge periods of shorter duration on the Inner route.

Figure 6.0 Mean total duration and number of bridge periods per assignment, by shipping route

Table 6.0 Analysis of the total duration, number and duration of each bridge periods per assignment, by shipping route (1)

Effect	Post hoc results (2)	Mean (sem)	F Statistics	p-value
Total duration (hours) of bridge periods per assignment Inner Route Hydrographers Passage GNE Channel	1 2 2	21.81 (2.95) 7.45 (0.82) 8.90 (0.68)	7.36	= 0.001
Number of bridge periods per assignment Inner Route Hydrographers Passage GNE Channel	1 2 2	7.51 (0.16) 1.14 (0.07) 1.33 (0.11)	263.27	< 0.001
Average duration (hours) of bridge periods Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	2.66 (0.36) 6.63 (0.80) 6.67 (0.58)	2.14	= 0.184

1. Results of full two-way Analysis of Variance (ANOVA) model. p < 0.01 considered statistically significant.
2. Results of Tukey’s Studentised Range test for post-hoc differences (Type 1 Error Rate = .01 p < 0.01 for differences between shipping routes from post-hoc

n/a = Post hoc testing not performed when main effects not significant

6.2 Percent of Pilotage Time spent on the Bridge

A further question to be addressed in terms of potential fatigue was the proportion of pilotage time spent on the bridge itself.

As Figure 6.1 shows the percent of pilotage time on the bridge ranged from 52 percent on the Inner Route to 86 and 98 percent for Hydrographers Passage and GNE Channel assignments, respectively. Statistical analysis revealed that the percentage of pilotage time on the bridge was significantly shorter for the Inner Route, with the percentage similar for the other shorter routes (Table 6.1). Thus on the shorter routes pilots spend nearly all of the pilotage time on the bridge.

Figure 6.1 Mean percent of pilotage time spent on the bridge, by shipping route

Table 6.1 Analysis of the percent of pilotage time spent on the bridge, by shipping route (1)

Effect	Post hoc results (2)	Mean (sem)	F Statistics	p-value
Percent of pilotage time on bridge Inner Route Hydrographers Passage GNE Channel	1 2 2	51.63 (5.04) 86.54(9.82) 98.34 (3.49)	8.57	=0.003

1. Results of full two-way Analysis of Variance (ANOVA) model. p < 0.01 considered statistically significant.
2. Results of Tukey’s Studentised Range test for post-hoc differences (Type 1 Error Rate = .01) p < 0.01 for differences between shipping routes from post-hoc

6.3 Percent of Bridgework at night and during critical hours

Earlier findings indicated that approximately 50 percent of ship time occurred during night hours. This figure was not unexpected given that starting times of assignments are distributed across the entire 24 hour period (Parker et al., Report No 2, 1998).

Logbook entries indicated that the percent of bridge time at night ranged from 37 percent for Hydrographers Passage, to 47 and 52 percent for the Inner Route and Hydrographers Passage, respectively (Figure 6.2). The absence of a significant shipping route effect is most likely related to the large standard errors associated with the mean values for Hydrographers Passage and the GNE Channel (Table 6.2).

The percent of bridge time during critical hours ranged from 28 percent for assignments on Hydrographers Passage to 29 percent for those on the Inner Route and GNE Channel (Figure 6.2). Thus, approximately one third of bridge time across all shipping routes occurred during time periods which are associated with decreases in alertness.

Figure 6.2 Mean percent of bridge time at night and during critical hours (2300-0600), by shipping route.

Table 6.2 Analysis of the percent of bridge time at night and during critical hours (2300-0600), by shipping route (1)

Effect	Post hoc results	Mean (sem)	F Statistics	p-value
Percent of bridge time at night (1818-0525) Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	46.91 (1.44) 37.28 (6.08) 52.47 (8.72)	0.27	=0.766
Percent of bridge periods in critical hours (2300-0600) Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	28.79 (5.63) 25.74(1.31) 28.96(6.56)	0.02	=0.982

1. Results of full two-way Analysis of Variance (ANOVA) model. p < 0.01 considered statistically significant. n/a = Post hoc testing not performed when main effects not significant

6.4 Sea conditions during Bridge Periods

Sea conditions were identified by the pilot advisory group as an important factor in determining the degree of difficulty associated with a particular assignment and with the potential to generate additional fatigue. To explore this issue respondents were asked to rate the sea conditions on the dimensions of calm (0-1.5 m), medium (1.5-5.4 m ) or heavy (5.4-13.7 m). The data is based on records taken during 1074 bridge periods undertaken during the months of May and June, 1998. With the exception of the GNE Channel it is unlikely that the sea conditions would have contributed to additional stress and/or fatigue during bridgework. As shown in Figure 6.3, sea conditions for the Inner route and Hydrographers Passage were calm during more than 50 percent of bridge periods, with medium conditions experienced about 30 percent of the time. In contrast, in the GNE Channel, sea conditions were medium during 69 percent of bridge periods with calm conditions during 32 percent of bridge time. This difference was not unexpected given the relative exposure of this region compared with the more protected routes by virtue of their position within the Barrier Reef. On all shipping routes there was an absence of heavy sea conditions during the sampling period. Weather conditions may have been a more significant factor in fatigue had the data collection period coincided with the cyclone season between December and March.

Figure 6.3 Mean percent of bridge periods associated with calm, medium and heavy sea conditions, by shipping route.

7.0 Sleep at Sea

An earlier investigation involving GBR pilots revealed that the sleep of pilots at sea is highly compromised in terms of its quantity and quality (Parker et al., 1997). It was suggested that the poor sleep patterns reflected the irregularity of work schedules and the on call nature of a pilots work. The strong relationship which exists between sleep deficiencies and fatigue prompted a closer examination of the nature and extent of sleep during work assignments at sea. Results shown in the earlier sections of this report indicated that opportunities for sleep at sea are determined by work patterns on the bridge. This section provides further insight into sleep patterns at sea with reference to sleep quality and the relationship to measures of fatigue and potential recuperative value.

7.1 Total Time in bed and Total Sleep Time per 24 hours

As shown in Figure 7.0 when in bed, pilots were mostly asleep, however the duration of time in bed during a 24 hour period differed significantly across the three shipping routes (Table 7.0).

Similarly, pilots total sleep time per 24 hours displayed a shipping route effect with pilots averaging less sleep on the GNE Channel than on the other two routes (Table 7.0). When evaluated per 24 hours, pilots averaged one hour of sleep on the GNE Channel and 4 and 5 hours for Hydrographers Passage and the Inner Route, respectively. Logbook entries for sleep times per 24 hours for Hydrographers Passage and the Inner Route were very consistent with questionnaire responses in which pilots indicated they averaged about 5 hours sleep per day while working at sea (Parker et al., Report No 4, 1998). In terms of the GNE Channel, the average sleep per 24 hours was consistent with data shown previously (Section 6.2, Table 6.1) which indicated pilots spend almost all of the pilotage time on the bridge in this region.

Figure 7.0 Mean total time in bed and total sleep time per 24 hours at sea, by shipping route

Table 7.0 Analysis of the total time in bed and total sleep time per 24 hours at sea, by shipping route (1)

Effect	Post hoc results (2)	Mean (sem)	F Statistics	p-value
Total time (hours) in bed per 24 hours Inner Route Hydrographers Passage GNE Channel	1 2 3	6.19 (0.21) 4.06 (0.79) 0.75 (0.30)	29.51	<0.001
Total sleep time (hours) per 24 hours Inner Route Hydrographers Passage GNE Channel	1 1 2	5.25 (0.25) 4.00 (0.74) 0.59 (0.30)	18.60	<0.001

1. Results of full two-way Analysis of Variance (ANOVA) model. p < 0.01 considered statistically significant.
2. Results of Tukey’s Studentised Range test for post-hoc differences (Type 1 Error Rate = .01) p < 0.01 for differences between shipping routes from post-hoc

7.2 Duration of each Sleep Period and Number of Sleep Periods per 24 hours

Figure 7.1 reveals that sleep on the Inner Route was taken in multiple sleep periods, in contrast with the other routes when sleep was experienced in a single sleep period. As Table 7.1 shows there was a significant shipping route effect for the sleep period duration and the number of sleep periods per 24 hours. Each sleep period was significantly longer (4.6 hours) on Hydrographers Passage, than on either the Inner Route (1.97 hours) or the GNE Channel (1.03 hours). The number of sleep periods were significantly greater on the Inner Route than the other two routes (Table 7.1).

That sleep was taken during three separate periods on the Inner Route is consistent with the assignment duration and multiple bridge periods on this route.

Figure 7.1 Mean duration of each sleep period and number of sleep periods per 24 hours at sea, by shipping route.

Table 7.1 Analysis of the duration of each sleep period and the number of sleep periods per 24 hours at sea, by shipping route (1)

Effect	Post hoc results (2)	Mean (sem)	F Statistics	p-value
Duration (hours) of each sleep period Inner Route Hydrographers Passage GNE Channel	1 2 1	1.97 (0.07) 4.66 (0.68 1.03 (0.21)	16.47	<0.001
No. of sleep periods per 24 hours Inner Route Hydrographers Passage GNE Channel	1 2 2	3.25 (0.12) 0.65 (0.09) 0.79 (0.36)	38.42	< 0.001

1. Results of full two-way Analysis of Variance (ANOVA) model. p < 0.01 considered statistically significant.
2. Results of Tukey’s Studentised Range test for post-hoc differences (Type 1 Error Rate = .01); p < 0.01 for differences between shipping routes from post-hoc

7.3 Other Sleep Characteristics associated with Fatigue

Characteristics of sleep such as the time taken to fall asleep (sleep latency) and that taken to feel alert following sleep (sleep inertia) can also provide an indication of fatigue levels. Sleep latency has been shown to decrease following periods of restricted sleep, with a latency of 5 minutes or less being associated with a fatigued state. Similarly, the time taken to feel alert after waking tends to increase as a person becomes more fatigued (Roth et al., 1989).

There was no significant shipping route effect for either measure (Table 7.2). Sleep latency time ranged from 4 minutes on the GNE Channel assignments to 6 and 7 minutes for Hydrographers Passage and the Inner Route, respectively. Thus the present sleep latency data was within or close to the classification of a fatigued state (Roth et al., 1989).

Table 7.2 also shows the percentage of sleep periods on each of the shipping routes with latencies of less than 5 minutes. Thirty eight and 31 percent of sleep periods met this criteria during assignments on the Inner Route and Hydrographers Passage, respectively. On GNE Channel assignments, 83 percent of sleep periods displayed sleep latencies of this duration. These findings suggested that the percent of sleep at sea with latencies < 5 minutes was higher for the GNE Channel, however the small number of sleep periods sampled on this route prevented statistical confirmation of this finding.

Sleep inertia represents a period of potential performance impairment that occurs immediately on waking. The time to feel alert ranged from 0.55 mins on the GNE Channel to 2 and 6 minutes for the Inner Route and Hydrographers Passage assignments, respectively and was not different between the shipping routes (Table 7.2). The short time taken to feel alert may be related to the on call nature of pilotage work and the associated apprehension which may counteract or mask any symptoms of sleep deprivation. Sleep inertia has been shown to be more severe for those in sleep deprived situations (Dinges, 1992), and continued sleepiness in non-sleep deprived subjects has been shown to last for 15 minutes with this time increasing in sleep deprived individuals (Haslam, 1982).

Sleep efficiency, defined as the percent of time in bed asleep was relatively high across all shipping routes (Table 7.2). This factor, coupled with the reduced sleep latency and little loss of sleep due to awakenings (Table 7.2), is consistent with a fatigued state.

Table 7.2 Analysis of other sleep characteristics associated with fatigue at sea, by shipping route (1)

Effect	Post hoc results	Mean (sem)	F Statistics	p-value
Sleep latency (mins) Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	7.23 (0.30) 6.36 (1.50) 3.82 (1.56)	0.13	=0.877
Time (mins) to feel alert Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	2.28 (0.32) 6.14 (2.52) 0.55 (0.18)	1.85	=0.158
Percent of sleep periods with sleep latency < 5 mins Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	38.88 (0.02) 31.82 (0.10) 83.33 (0.16)	0.61	=0.542
Time (mins) lost due to awakenings Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	3.20 (0.38) 3.05 (1.67) 1.92 (0.86)	0.01	= 0.998
Sleep efficiency Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	90.25 (3.79) 92.82 (1.71) 93.61 (3.55)	0.02	=0.998

1. Results of full two-way Analysis of Variance (ANOVA) model. p < 0.01 considered statistically significant. n/a = Post hoc testing not performed when main effects not significant

7.4 Timing of Sleep

Bridgework at sea determines not only the duration but also the timing of sleep. Sleep during physiologically appropriate times (2200-0800 hours) has the potential to be of greater recuperative value (Akerstedt, 1995; Kecklund et al., 1997).

7.4.1 Percent of sleep within 2200-0800 hours

As Figure 7.3 shows the percentage of sleep within these hours ranged from 50 to 81percent. A significantly higher percentage of sleep was taken within these hours on Hydrographers Passage (81 percent) compared with the Inner Route (50 percent), and Great North East Channel (59 percent) (Table 7.3). This finding reflects the need for shipping schedules on Hydrographers passage to coincide with port facilities and is consistent with the single bridge and sleep period reported previously. In contrast, the longer duration of Inner Route assignments and the greater number of bridge and sleep periods resulted in approximately half the sleep being taken outside of this optimal time period.

That 50 and 40 percent of sleep occurred outside optimal hours for the Inner Route and GNE Channel, respectively, raises further questions regarding the recuperative value of sleep of 1-2 hour duration to counteract fatigue while on the bridge in these regions, particularly on the longer Inner Route.

As the frequency distribution in Figure 7.4 shows approximately 20 percent of sleep starting times occurred across most of the hours in the 24 hour cycle (between midnight and 2200) which indicates that pilots are frequently required to sleep at sub-optimal times. A peak in sleep starting times occurred between the hours of 2000 and midnight on the three routes. However, previous sections have shown that it is likely that sleep beginning between these hours is inclined to be fragmented, particularly on the Inner Route.

Figure 7.2 Mean percent of sleep periods inside 2200-0800 at sea, by shipping route.

Table 7.3 Analysis of the percent of sleep periods inside 2200-0800 at sea, by shipping route (1)

Effect	Post hoc results (2)	Mean (sem)	F Statistics	p-value
Percent of sleep inside 2200-0800 Inner Route Hydrographers Passage GNE Channel	1 2 1	50.18 (1.77) 81.23 (8.37 59.99 (24.49)	25.31	<0.001

1. Results of full two-way Analysis of Variance (ANOVA) model. p < 0.01 considered statistically significant.
2. Results of Tukey’s Studentised Range test for post-hoc differences (Type 1 Error Rate = .01) p < 0.01 for differences between shipping routes from post-hoc

Figure 7.3 Frequency distribution - starting times of sleep periods at sea, by shipping route.

7.5 Sleep Quality

Earlier sections of this report have indicated that sleep at sea is of short duration, sometimes fragmented and characterised by factors associated with fatigue such as short latency. Moreover, up to 50 percent of this sleep is outside optimal sleeping hours. This section includes measures which are indicative of the quality of sleep and represent the pilots perception with respect to; the ease of falling asleep, ease of arising, sufficiency of sleep, and feeling rested and depth of sleep. The sleep quality measure was determined from the sum of responses to scales for these measures (range 5-25), a higher score indicating better quality sleep.

Ratings of sleep quality were at the higher end of the sleep quality range and were similar across the different shipping routes (Figure 7.5) and (Table 7.4). This result suggests that the high sleep quality scores may indicate that the short and fragmented sleep at sea provided some restorative value. However, these ratings were inconsistent with previous sleep quality data on this group which showed that the majority of pilots experienced poor-very poor sleep (Parker et al., 1997). This discrepancy suggests that the pilots may have slightly overrated their sleep which is common when subjective assessments are used, particularly in occupational settings.

Figure 7.4 Sleep quality ratings at sea, by shipping route

Table 7.4 Analysis of sleep quality ratings at sea, by shipping route (1)

Effect	Post hoc results	Mean (sem)	F Statistics	p-value
Rating of sleep quality Inner Route Hydrographers Passage GNE Channel	n/a n/a n/a	20.08 (0.14) 21.00(0.90) 19.17(1.76)	2.39	=0.092

1. Results of full two-way Analysis of Variance (ANOVA) model. p < 0.01 considered statistically significant. n/a = Post hoc testing not performed when main effects not significant

7.6 Sleep Debt

Sleep debt is defined as the difference in the duration of home sleep and sleep in another location. As Table 7.5 shows, earlier estimates of sleep debt at sea in this population indicated a daily debt of 2.3 hours (Parker et al., Report No 4, 1998).

When compared with sleep ashore during assignment breaks, logbooks indicated pilots experienced a daily sleep debt of 5.7 and 6 hours for the Inner Route and Hydrographers Passage, respectively. (Table 7.6). When compared with sleep at home, logbook daily sleep debt for both these shipping routes was approximately 3 hours. Hence, whether sleep debt at sea is based on at home, or ashore sleep hours it is a greater problem on longer Inner Route assignments where the sleep debt is incurred over several days.

Table 7.5 Questionnaire responses: sleep patterns at home, ashore and at sea.

Mean sleep duration (hours) per 24 hours at Home	Mean sleep duration (hours) per 24 hours ashore	Mean sleep duration (hours) per 24 hours at sea	Daily sleep debt (hours)
7.8 (0.29)	8.0 (0.35)	5.5 (0.38)	2.3 hours (home –sea)

Table 7.6 Logbook responses: sleep patterns at sea and ashore, by shipping route.

Sleep duration per 24 hours	Inner Route	Hydrographers Passage
Total sleep time (hours) ashore before assignments on shipping routes	10.99	10.06
Total sleep time at sea	5.25	4.00
Sleep debt per 24 hours	5.74	6.06

7.7 Sea conditions during sleep

The results indicated that it was unlikely that sleep at sea was affected to any degree by sea conditions. For instance, as indicated by Figure 7.4 on the Inner Route approximately 50 percent of sleep periods were associated with calm or medium sea conditions. During Hydrographers Passage pilotages 90 percent of sleep was associated with calm conditions. In contrast, all sleep periods on the GNE Channel assignments were associated with medium sea conditions. That sea conditions in this latter region during the more stable winter months were medium is consistent with the relative exposure of the region to prevailing weather patterns compared with the other shipping routes.

Figure 7.5 Mean percent of sleep periods with calm, medium or heavy sea conditions, by shipping route

[back to top]