Shipping Safety

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

Codes Manuals and Reports

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

Section 1.0 Background

A recent survey of the lifestyle behaviours and industry specific factors associated with the health, stress and fatigue of Australian seafarers identified several areas which may impinge upon the health and potential for fatigue in GBR pilots (Parker et al., 1997). An important finding was the poor quality and short duration of sleep experienced by pilots at sea which has been shown in other populations to be associated with mood and performance decrements and increased fatigue. While fatigue has clearly been identified as a major factor in accident risk in other transport industries it has only recently been shown that fatigue may be a contributing factor to a significantly greater number of marine accidents than previously reported (McCallum et al., 1996). The nature and demands of marine pilotage in the Great Barrier Reef and the environmental sensitivity of this region was the catalyst for the commissioning by AMSA of an investigation into the work practices of GBR pilots and the likely impact of these work practices on the fatigue of pilots. The project was undertaken by a research team coordinated by the School of Human Movement Studies at the Queensland University of Technology and involved the following phases:

The use of on-tour logs to provide specific information about ship time (bridge and sleep periods) and non-ship time ashore (sleep periods). This information will be used to help validate work schedule data and provide information on the relationship between work/rest patterns and alertness.

The information resulting from these investigations is both comprehensive and complementary and will be applied to the development of strategies and guidelines to reduce the potential for fatigue and risk of accident.

A Pilot Advisory Group, group comprising one member from each of the pilotage companies operating in the Great Barrier Reef – Torres Strait region was formed to provide industry specific direction to the research. This group was involved in the design and validation of the survey instruments and assisted in gaining and maintaining the support and cooperation of the pilot companies and their members. Regular meetings were held with the researchers and representatives from AMSA and there were frequent informal communications between members of the Advisory Group and the research team.

This report primarily presents the results from the on-tour log-book analyses, however, reference is also made to information derived from earlier stages of the research program. The various phases of the program are designed to provide a progressive and complementary analysis of the work practices of GBR pilots using a variety of research strategies. The findings have been combined to identify those aspects of pilotage which have the potential for fatigue and to provide a number of recommendations designed to implement appropriate fatigue management procedures.

Section 2.0 Introduction

The Great Barrier Reef region is recognised both nationally and internationally as a unique and environmentally sensitive area. The region is registered on the World Heritage List and in 1990, was identified by the International Maritime Organisation as the world’s first, and to date only, ‘Particularly Sensitive Sea Area’ (Queensland Department of Transport, 1996; AMSA, 1996). Coastal pilotage in the region is centred on the movement of ships along and through the Great Barrier Reef and into the Torres Strait region, with pilotage assignments ranging from 12 to 60 hours. Pilots often experience testing conditions and have to cope with varying ship, equipment and crew standards (90 percent of the piloted vessels are foreign flag), combined with the vagaries of weather and recreational traffic. Additionally, ships may be carrying potentially hazardous and highly polluting cargo, and are often deep laden causing intricate navigational problems. Extrinsic factors such as the commercial aspects of pilotage appear to have caused considerable additional stress to the pilotage workforce. Regulatory changes to pilotage in 1993 saw the emergence of three competing pilotage companies in the region replacing the 100 year old one provider situation. Hence, it is possible that the characteristics of the work patterns of GBR pilots and a range of extrinsic factors may have the potential to impact on their potential for fatigue and work performance.

Earlier research indicates that a pilots work is generally characterised by irregular scheduling and compromised sleep. For example, work schedule data from United Kingdom pilots (Shipley & Cook, 1980) and the present pilots revealed that the entire 24 hour period was represented and assignment and break starting times commonly occurred outside of the normal circadian cycle. Night work was common and findings from the present investigation indicated that approximately 50 percent of ship time of GBR pilots is undertaken during the night (where night is defined as between 1818 and 0525 hours) (Parker et al., Report No 2, 1998). Compromised sleep at sea has been consistently identified in studies from pilotage groups in the United Kingdom (Shipley & Cook, 1980), the Netherlands (de Vries-Grier, 1982), the Port Phillip region of Australia (Berger, 1984) and the Great Barrier Reef region (Parker et al., 1997). Thus, it would appear that work and sleep patterns in marine pilotage operations are conducive to workplace fatigue. Maritime research has shown that fragmented sleep periods, sleeping at physiologically inappropriate times of day and insufficient time between work periods to obtain restorative sleep were critical factors contributing to mariner fatigue (Sanquist et al., 1996).

Irregular working hours, particularly when involving the whole 24 hour cycle, require working at times which may oppose the natural diurnal biological rhythms of the human body. In this situation sleep quantity and quality are frequently compromised and fatigue occurs. If the fatigue is excessive over the shorter or longer term, decrements in performance may result particularly in cognitive, vigilance and memory tasks (Akerstedt, 1995; Condon et al., 1988; Krueger, 1989; Monk, 1989; Monk & Folkard, 1992; Rosekind et al., 1996; Scott & Ladou, 1990; Transportation Safety Board (TSB), 1997; Griffiths, 1993; Krueger, 1989). The increased accident risk during the early morning hours, particularly 0300 hours, with an additional but smaller peak in accident risk during the mid afternoon has been attributed to working during the circadian trough in alertness (Summala & Mikkola, 1994; Brown, 1994; Couper, 1996; Folkard, 1997; Sanquist et al., 1996; Mitler et al., 1988; Hopkins 1992;). In terms of pilotage performance, this may result in lack of awareness of position and monitoring of navigational equipment, failure to respond quickly to adverse situations, inaccurate calculations, impaired judgment and over-reliance on radar and automated equipment (Couper, 1996; Dinges, 1992; Sanquist et al., 1996).

While in the past, only a small proportion of marine accidents were thought to be fatigue-related, more recent reports from the US and Japan have suggested otherwise. For instance, studies of accidents in US waters have indicated a 10-fold increase in the fatigue contribution to critical vessel and personnel injury casualties since 1993 (McCallum et al., 1996). Japanese investigators found that ‘lack of alertness’ and ‘dozing during navigation’ accounted for a significant proportion of groundings, strandings and collisions occurring between 1985 and 1991 (Japan Maritime Research Institute, 1993). Additionally, official Australian statistics revealed 9.2 percent of shipping casualties occurring in Australian waters between January 1994 and January 1998 were fatigue-related. However, one author has suggested that a figure closer to 30 percent would be more realistic when performance impairments due to chronic fatigue are considered (Filor, 1998).

During the present project the approach to exploring fatigue was based on examination of factors identified as potential contributors to stress and fatigue such as irregular work and sleep patterns, and their association with changes in alertness. As self-report measures of alertness have been shown to have a strong association with performance (Gillberg et al., 1994) changes in alertness over a number of bridge assignments were recorded to evaluate the likely impact of work practices on fatigue and work performance. While previous research and retrospective analysis of work schedule data (Parker et al., Report No 2, 1998) has identified factors which potentiate fatigue, more detailed analyses were required to determine the specific details of work and rest at sea and their association with alertness and fatigue. This data will serve to validate earlier data and provide a more complete picture of the work practices of these pilots working under unique conditions and in a very significant region.

Specifically, the final phase of the project was designed to examine:

Section 3.0 Methodology

Development and implementation of the methodology for this phase of the research program involved primarily the selection of an appropriate instrument to allow pilots to record specific details of work assignments and breaks between these assignments. In this selection process it was necessary to select a valid instrument that could be administered by pilots with minimum disruption to their work. For this purpose a log book used in an earlier study of the work practices of watchkeepers in the US Coast Guard (Sanquist, 1996) and the UK (Seafarers International Research Centre, 1996) was used. The pilot advisory group provided suggestions which were helpful in making modifications to the logbook to improve its relevance to the work of GBR pilots.

3.1 Sample

Logbooks were distributed to 58 GBR pilots. The sample represented the three pilotage companies operating in the Great Barrier Reef region. The two larger companies each employ 29 and 26 pilots, respectively, with 4 pilots employed by the smaller company.

3.2 Log Books

Pocket sized on tour logbooks were designed to: (i) provide specific information on bridgework such as the duration, timing, alertness ratings and fatigue symptoms; (ii) assess the duration, timing and quality of sleep periods at sea and ashore between assignments; (iii) validate the responses from the questionnaires (Parker et al., Report No 4, 1998) and the analysis of ship and non-ship time from the work schedules (Parker et al., Report No 2, 1998); and (iv) identify factors associated with high fatigue and stress and low alertness on the bridge. Pilots were requested to complete the logbooks after each work and sleep period for consecutive work assignments during a tour of duty.

[back to top]

3.3 Log Book distribution and reminder system

Eight logbooks (including a stamped addressed envelope for returns) were distributed by the pilot companies and posted to the home address of each pilot. An information package with the logbooks provided detailed instructions on completing the logs and addressed issues relating to confidentiality and the anonymity of data presentation. A reminder system encouraging project participation consisted of two individual reminders to pilots and facsimile messages to pilot accommodation on Thursday Island.

3.4 Logbook Development

As indicated earlier, a previously validated logbook was modified for implementation in this investigation (Sanquist et al., 1996; Seafarers International Research Centre, 1996) and reviewed by the Pilot Advisory Group to ensure measures specific to Reef pilotage were included. In addition to the pilot advisory group the contents and organisation of the logbook were reviewed by Professor Simon Folkard (Body Rhythms and Shiftwork Centre, University of Wales, Swansea) to ensure the inclusion of key items related to sleep and circadian issues. The main changes to the logbooks included: (i) an assessment of the duration of pilotage time during an assignment; (ii) travel time to and from the ship; (iii) ratings of stress and fatigue levels for an assignment; and (iv) ratings of the degree to which a large number of industry and pilotage specific factors contributed to assignment stress and/or fatigue.

3.5 Pilot Testing of Log Books

The logbooks were tested by a pilot during a work assignment and comments regarding content, suitability and acceptability were provided. These comments were incorporated into the logbooks prior to distribution following consultation with the Pilot Advisory Group. As the original logbooks had been previously used in maritime settings further trialing was not considered necessary. A copy of the logbook content is shown in Appendix 3.

3.6 Measures

3.6.1 Work assignment characteristics

Items in this section were designed to provide a general overview of the assignment and were completed once only. Questions related to: (i) the shipping route; (ii) date, time and location of embarkation and disembarkation; (iii) date and time pilot duties commenced; (iv) date and time pilot duties ceased; (v) travel time to and from boarding grounds.

3.6.2 Work periods on bridge

For each work period on the bridge items in this section were completed. These related to: (I) sea conditions; (ii) date and time of each bridge period; (iii) alertness ratings at the beginning middle and end of a bridge period; and (iii) symptoms of fatigue experienced during bridgework.

Respondents were asked to rate their alertness at the beginning, middle and end of bridge periods on a Visual Analogue Scale (VAS), (range ‘very sleepy’ to ‘very alert’). Alertness values were scored on a 1-9 scale based on the position indicated on the scale.

The dimensions of fatigue symptoms consisted of: (i) forgetfulness; (ii) distractedness; (iii) attention difficulties; (iv) decreased motivation; (v) did things at wrong time; (vi) sore muscles; (vii) heavy eyelids; (viii) desire to sit or lay down; (ix) itchy eyes; and (x) difficulty focusing eyes. Respondents were asked to rate the frequency of experiencing fatigue symptoms. Items were scored on a five-point Likert scale ranging from ‘not at all’ (Scale =1) to ‘very much’ (Scale = 7) with a single total score calculated (range 10-70). A higher score was associated with a greater frequency of symptoms.

3.6.3 Sleep periods at sea

Items in this section were completed for each sleep period at sea. Items related to: (i) sea conditions; (ii) date and time of sleep periods; (iii) duration of sleep periods; (iv) time to fall asleep; (v) number of awakenings; and (vi) ratings of sleep quality. Sleep quality was assessed on the dimensions of ease of falling asleep, ease of rising, sleep period sufficiency, restedness and sleep depth. Respondents were asked to rate sleep quality on a five-point Likert scale ranging from ‘least’ (Scale =1) to ‘most’ (Scale = 5), with a single total score calculated (range 5-25). A higher score indicated greater sleep quality.

3.6.4 End of work assignment assessment

This section was completed once only, at the end of a work assignment at sea. Items in this section were designed to examine the overall stress and fatigue of the work assignment. Respondents were asked to rate their stress and fatigue levels for the assignment on a Visual Analogue Scale (VAS), (range ‘very sleepy’ to ‘very alert’). Stress and fatigue was scored on a 1-100 scale based on the position indicated on the scale.

Additionally, this section assessed the degree to which industry specific factors increased overall stress and/or fatigue. Dimensions of industry specific factors related to: (i) bridge team; (ii) ship length; (iii) ship speed; (iv) cargo type; (v) equipment availability; (vi) ship handling capabilities; (vii) under keel clearance; (viii) weather; (ix) visibility; (x) shipping traffic; (xi) recreation traffic; (xii) language communication difficulties; (xiii) meals; (xiv) accommodation; (xv) travel to the boarding ground; and (xvi) travel from the boarding ground. Respondents were asked to rate the degree of contribution of these factors on a Visual Analogue Scale (range ‘not at all’ to ‘very much’). On the basis of factor analysis three subscales were identified. These were: ship and environment factors, onboard factors and travel factors.

3.6.5 Additional comments

A separate page was provided to record any additional comments relating to issues not addressed in the logbook and to expand on problems encountered during particular assignments.

3.6.6 Sleep periods ashore

This section was completed each time pilots slept ashore between work assignments and was recorded for up to three days. The issues covered were identical to those addressed in the sleep periods at sea.

3.7 Data Analysis

3.7.1 Quantitative Data

The quantitative data was analysed using SAS-PC and standard univariate statistics were used to describe responses. Repeated measures Analysis of Variance (ANOVA) was undertaken to compare mean aggregate scores for logbook variables on the Inner Route, Hydrographers Passage and the GNE Channel. When there was significant evidence of differences, post-hoc tests were used to clarify the nature of those differences.

Some modification of existing scales was undertaken to improve reliability and validity of responses. However, in all cases, the aggregate scores were re-scaled so that they had the same maximum and minimum as the original scales enabling comparisons with data from other occupational groups to be made (Barton et al., 1995; Sanquist et al., 1996).

Factor analysis, using the principal components method to extract the initial factors and a promax (oblique) rotation method, was used to help identify constructs underlying a series of questions dealing with factors contributing to work assignment stress and fatigue. Further item analysis involving assessment of the reliability of scales and their interpretability, was undertaken before the scales were finalised. Items included in the sub scales are shown in Section 3.6.4.

Cronbach’s alpha coefficients were calculated to determine the reliability of each of the aggregate scores used in the report. Scales showed a high level of reliability with Cronbach’s alpha coefficients for travel factors, onboard factors and ship and environment factors being 0.73, 0.87 and 0.90, respectively.

A cut-off of 0.01 was used to assess the statistical significance of p-values in order to exercise some control over the overall experimentwise error rate (type I error rate) and minimise the number of spuriously significant results.

A series of multiple linear regression models were used to assess contributions to variation in mean overall fatigue, stress, and minimum alertness measures of pilot data reflecting ashore and at-sea variables. After accounting for the variation in fatigue, stress, and alertness among the pilots, six incremental levels of modelling were attempted, each level evaluated a set of data at a higher level of data refinement, considering the percentage of additional variation in fatigue, stress, and alertness accounted for by:

  1. the duration of the break preceding an assignment;
  2. travel duration to and from the ship per assignment;
  3. work variables, including the duration of the work assignment, duration of pilotage duties, duration of time on the bridge, average duration of bridge periods per assignment, number of bridge periods per 24 hour period during the assignment; and number of critical hours spent on the bridge
  4. total duration of time in bed per 24 hours at sea;
  5. total duration of sleep during the break preceding the assignment; and
  6. total duration of non critical sleep during a break.

These models were considered for the Inner Route only, as there were insufficient work and sleep periods for meaningful analysis of the other two routes. R2 values and changes in R2 values across the incremental models, expressed as percentages of variation in fatigue, stress, and alertness accounted for by the variables in the model, were used to summarise the results.

3.7.2 Qualitative Data

Participants were requested to include any additional comments on the issues covered within the logbooks or on topics that were omitted from the logbooks. These comments facilitated the interpretation of the logbook data given the large variation in work assignments, and provided a clearer understanding of the context within which responses to the logbooks were made.

This qualitative data was analysed by examining the comments for recurrent patterns and themes. The data has been incorporated within the relevant sections of the results.

3.8 Presentation of Results

Results of the logbook analysis in the present report are presented in four key sections: (i) patterns of activity on the ship (bridgework and sleep); (ii) patterns of activity during breaks (sleep); (iii) measures of alertness and fatigue on the bridge; (iv) factors contributing to fatigue, stress and alertness on the bridge. Figure 3.0 shows a schematic view of the presentation of results.

Figure 3.0 Schematic diagram of key measures from the logbooks

wpe2.jpg (41475 bytes)

[back to top]

last updated: October 1998