i
Southampton Solent University
Warsash Maritime Academy Maritime and Technology Faculty
MSc in Shipping Operations
A CRITICAL ANALYSIS OF
HARBOUR TOWAGE
OPERATIONS RISKS TO
SAFETY.
Stephen Ford
31st
May 2013
ii
I. Abstract
Recent fatal harbour towage operational accidents highlight a potential safety concern. Initial investigation revealed a lack of empirical scientific research evidence regarding specific risks encountered during harbour towage operations. This project therefore sought to establish if harbour towage operations face particular risks. Comparison of harbour and non harbour towage operations indicated different risk profiles, with harbour towage accidents more likely to result in a collision involving loss of life. Statistical comparison of harbour towage and non harbour towage risk factors corroborated this. Certain risk factors were only present in harbour towage operations, and risk factor volumes were greater. Statistical testing of the relationship between individual risk factor and consequence significance also revealed a link. A number of risk factors were identified as influential in harbour towage operations; these ranged from safety management systems, tow planning, and interaction, to vessel speed, training and tug type. The research method employed, combining quantitative and qualitative surveying through triangulation, to objectively analyse and compare the data, demonstrated a degree of success; although a longitudinal methodology might better align frequency with likelihood, and better enable measurement of success of any intervention. The findings suggest a number of recommendations including, improved confidential hazardous event reporting, legislative reform to establish equitable regulatory oversight & monitoring, and enhanced training provision for individuals involved in harbour towage operations.
iii
Contents
Page
I. Abstract iv
Contents v List of tables & graphs vi
List of abbreviations viii
1.0 Literature Review 1 2.0 Methodology 6
2.1 Secondary data 8 2.2 Case Studies 9 2.3 Primary Data: Questionnaire 11 2.4 Grounded Theory 15
3.0 Results 18
3.1 Case Study Account 18 3.2 Questionnaire Account 22 3.3 Expert Interview Account 28 3.4 Comparison of Harbour Towage and Non Harbour Towage data 32 3.5 Hypothesis Testing: Chi Square test 37 3.6 Comparison of quantitative data 38 3.7 Risk Factors 45 3.8 Relationship between Risk Factor and Consequence Severity 48
4.0 Discussion 51
4.1 Suitability of Method 51 4.2 Harbour Towage versus Non Harbour Towage 53 4.3 Case Study versus Questionnaire Data 54 4.4 Risk Factor Quantity 57 4.5 Risk Factor Severity 57 4.6 Individual Risk Factor Account 58
5.0 Conclusions 63
5.1 Suitability of method 63 5.2 Analysis of risks to safety 64
6.0 Recommendations 67
6.1 Harbour towage operations reporting and research 67 6.2 Equitable regulatory oversight and monitoring 67 6.3 Enhanced training provision 68
7.0 References 69
Appendix A. A A-1 Example Harbour Towage Safety Questionnaire A
iv
List of tables and graphs Page
Table 1: Research Project Programme 7 Table 2: Questionnaire Sections 11 Table 3: Publicity Programme 12 Table 4: Questionnaire Distribution Groups 13 Table 5: Distribution of Case Studies 18 Graph 1: Apportionment of Accident Type 18 Graph 2: Apportionment of Event Consequence 19 Graph 3: Distribution of Tug Type 19 Graph 4: Distribution of Tug Bollard Pull 19 Graph 5: Distribution of Tow Position 20 Graph 6: Towed Vessel Type 20 Graph 7: Bow Form Distribution 20 Graph 8: Distribution of Towed Vessels Size 21 Table 6: Questionnaire Source 22 Graph 9: Safety Occurrence Description 23 Graph 10: Safety Occurrence Potential Result 23 Graph 11: Safety Occurrence Potential Consequence 24 Graph 12: Distribution of Tug Type 24 Graph 13: Distribution of Tug power 24 Table 7: Tug Power Categories 24 Graph 14: Tow Operation Type Distribution 25 Graph 15: Towed Vessel Category Distribution 25 Graph 16: Towed Vessel Bow Form Distribution 26 Graph 17: Towed Vessel Size Category 26 Table 8: Distribution of Towed Vessel Deadweight Categories 26 Graph 18: Prevailing Environmental Conditions 27 Table 9: Evidence Category: Speed, Interaction & Girting 28 Table 10: Evidence Category: Extreme Conditions 28 Table 11: Evidence Category: Legislation 28 Table 12: Evidence Category: Tow Planning & Command 29 Table 13: Evidence Category: Maintenance 29 Table 14: Evidence Category: Design & Complexity 29 Table 15: Evidence Category: Seamanship & Rope Management 29 Table 16: Evidence Category: Crewing 30 Table 17: Evidence Category: Stability 30 Table 18: Evidence Category: Fatigue 30 Table 19: Evidence Category: Training 30 Table 20: Evidence Category: Personal Qualities & Negative Attitudes 31 Table 21: Evidence Category: Time 31 Graph 19: HT to NHT Incident Category Comparison 32 Graph 20: HT to NHT Comparison of Consequences 32 Graph 21: Comparison of HT and NHT Towage Position 33 Graph 22: Comparison of Towed Vessel Type between HT and NHT CS 33 Graph 23: Comparison between HT and NHT Bow Form Distribution 34 Graph 24: Comparison of Towed Vessel Size 34 Graph 25: Case Study Comparison of Risk Factor Frequencies 35 Graph 26: Comparison of HT and NHT Risk Factor Volumes 36 Table 22: Chi Square Test Values 37 Graph 27: Comparison of CS and QU Events 38 Graph 28: Comparison of Consequence Frequency 38 Graph 29: Comparison of Tug Type 39 Graph 30: Comparison of Tug Power 39 Graph 31: Comparison of Tow Position 40 Graph 32: Comparison of Towed Vessel Type 40 Graph 33: Towed Vessel Bow Form Comparison 41 Graph 34: Towed Vessel Size Comparison 41 Graph 35: Comparison of Harbour Towage Risk Factors 42
v
Graph 36: Rank Difference Case Study Versus Questionnaire 43 Graph 37: Additional Risk Factors identified in Case Studies 44 Graph 38: Comparison of Weighted and Un-weighted Risk Factors 45 Graph 39: Risk Factor Percentage Increase due to Perceived Importance 46 Graph 40: Risk Factor Rank Change due to Perceived Importance 47 Table 23: Pearson’s r Significant Number Values 48 Graph 41: Risk Factor Rank Movement 49 Graph 43: Risk Factor Frequency Variation 50 Table 24: Key Harbour Towage Risk Factors 65
vi
List of Abbreviations
ASD Azimuth Stern Drive [tug]
ATSB Australian Transport Safety Board
BTA British Tugowners Association
CS Case Study
DSB Dutch Safety Board
EI Expert Interview
EMSA European Maritime Safety Agency
HT Harbour Towage [operation]
ILO International Labour Organisation
IMO International Maritime Organisation
ITA International Tugmasters Association
KMSB Korean Maritime Safety Board
MAIB Marine Accident Investigation Branch
NHT Non Harbour Towage [operation]
QU Questionnaire
TSBC Transport Safety Board of Canada
UK United Kingdom
US United States of America (USA)
USCG United States Coastguard
VS Voith Schneider [tug]
vii
Glossary
Case Study Individual maritime safety agency accident report, analysed to extract harbour towage operations risk and safety data.
Expert Interview Qualitative interview (supported by written submission and
observational analysis) of expert witness experience of harbour towage operations safety.
Harbour Towage Movement, berthing or unberthing of a vessel with the assistance of a Operation tug(s) within a harbour, port or equivalent area.
Girting (Also similar: Girding and Tripping). Where a vessel is caused to
potentially capsize, most commonly as a result of external towline and
interaction forces.
Interaction Hydrodynamic forces commonly found immediately adjacent to a
vessel moving through the water.
Non Harbour Any operation or activity other than harbour towage, involving a tug(s) Towage Operation carried out in any sea area.
Questionnaire Survey of current practitioners experience of harbour towage
operations safety, using a Likert style questionnaire (See Appendix A).
Risk Factor An element whose presence or absence has potential to lead to lead
to an unsafe event.
Tow Planning Planning and management of a harbour towage operation, commonly
undertaken by a licensed pilot.
1
The Transport Safety Board of Canada (2013) report’s that, tugs & barges along with Bulk
carriers, ‘were involved most often in accidents’1 (thirteen percent). This includes having the
highest number of fatalities and the highest number of accidents aboard ship.
While it might be argued that this is due to a highly developed river tug sector in Canada, the
Australian Transport Safety Board (2011) shipping occurrence statistics showed that eight
percent of collisions, ten percent of contact damage and seventeen percent of capsizing,
involved tugs.
Well publicised, but tragic harbour towage accidents include:
19th July 2012, the Tug Madison was capsized by her tow, the dredge barge
Arthur J., on Lake Huron (USCG, 2012);
12th August 2011, the tug Chieftain capsized with the loss of one life, off Convoys
Wharf on the River Thames, while towing the crane barge Skyline (MAIB, 2012
A);
11th June 2011, the tug Adonis, while engaged in moving the barge Chrysus, in
Gladstone, was capsized with the loss of one life (ATSB, 2013);
11th November 2010, the tug Fairplay 22 capsized with the loss of two lives, in
gale force winds off the Hook of Holland, while making fast to the Stena
Britannica (DSB, 2010).
Henson (2012) points out that, ‘tug operations near the bow of a ship having headway are
very risky; the higher the ship’s speed, the larger the risks’. Dand (1975) reporting on ship
model tests said that, ‘interaction forces varied with the square of the speed; and near the
fore body of a ship the tug may drive itself under the bow’.
A Dutch Safety Board (2010) report into the fatal collision and capsize of the tug Fairplay 22,
concluded that, ‘the tug had sailed close to bulbous bow, and within the hydrodynamic
sphere of influence; here it was unable to maintain a safe distance and collided with vessel’.
1 Excludes Fishing Vessels.
1.0 Literature Review
2
The British Tug Association (2010a) described an example of the associated risk of Girting
where, ‘a tug acting as brake while assisting the berthing a barge started an uncontrolled
yaw, leading to her capsize and foundering’.
The New Zealand Transport Accident Investigation Commission (2001) report into the
capsize and sinking of the tug Nautilus III, point out the importance of adequate Tow
Planning in the prevention of harbour towage accidents.
MAIB (2012) reporting on the fatal accident to the Chiefton supported this view, pointing out
that ‘the passage plan centred almost entirely on the bridge transit phases and did not
properly consider the need for river passage planning or its related risks’.
Referring to the same accident, MAIB highlighted the importance of the development of
adequate Safety Management Systems. They concluded that, ‘there was no evidence that
the tug operating company had conducted formal risk assessments of their vessels’
operations; the watertight integrity discipline on board Chiefton was weak; and, the
functionality of Chiefton’s towing hook release system was in doubt’.
Kunze (2011) talking at the British Tugowners Association’s Safety Seminar, underlined the
importance of Training, highlighting how ship simulators can provide opportunity for tug
masters to, ‘gain competence and confidence’. The 2012 BTA Safety Seminar again
highlighted the importance of Training, ‘delegates citing continued instances of poor
seamanship observed onboard assisted vessels, leading to dangerous occurrences for tugs’.
Stockman (2010) in his report on the near girting and capsize of the tug Stockton II,
underlined that Following Operational Procedures can be critical to safety. He pointed to
video footage of the incident illustrating, ‘as the tug heeled over with its doors pinned open,
had the tow rope not parted, the results of the incident could have been far more serious’.
The USCG (2009) Marine Safety Information Bulletin, dealing with reducing ‘Downstreaming’
safety incidents, underlines the critical importance of adequate Tug Handling.
Lack of manoeuvring space is also a risk to safety during harbour towage operations, as
highlighted by BTA (2010b) in their description of a collision between a tug and a vessel
while operating in a narrow channel.
3
The European Harbour Masters committee (2010) identified concern over the issue of ship
size, stating that Pilots and tug captains are ‘increasingly facing operational problems
handling ever growing ship sizes’.
Henson (2011) in Safe Tug Procedures underlines the importance in choice of tug type, to
help ensure safety of operations. He points out, ‘if tugs with propulsion units aft are very
close to the ship’s bow, to get clear by steering away, the tug’s stern will come closer to the
ship, increasing the suction forces and consequently the risk of hitting the bow’.
The Australian Transport Safety Board (2006) identify risks from lack of maintenance in their
report (No. 224) into a collision between a bulk carrier and a tug. They conclude that, ‘a
crack in the tug’s starboard main engine clutch oil discharge pipe, led to the engine’s
shutdown; this caused the tug’s stern to swing sharply to starboard, making heavy contact
with the ship, and puncturing the ship’s shell plating’.
The New Zealand Transport Accident Investigation Commission (2000) underlined the
importance of communications in harbour towage operations safety. In their report into the
man overboard and near capsize on a tug, they identified the safety issue of, ‘poor
communication between bridge team and crew at mooring stations; and insufficient
communication between tug skipper and pilot’ leading to the unsafe situation.
Livingstone (2012) in The International Pilot, points out the risk that environmental conditions
can place on harbour towage safety. In the fatal accident when the tug Flying Phantom was
girted and sank, thick fog may have led to a disorientation of the tug crew.
A joint paper produced by the European Tugowners and the European Pilots Associations’
(2011) demonstrates how advances in vessel design may produce risks to harbour towage
safety. They point to the, ‘operational problems European pilots and tug operators have
increasingly experienced over the last decade relating to the type and strength of deck
equipment on board of ships’; this highlights a contradiction posed by increased tug bollard
pull, versus moderated bollard structural strength.
Legislation with respect to harbour towage operations is complex, tending to cascade from
IMO (2013) International Conventions. Principle treaties include, the International
Convention for the Safety of Life at Sea (SOLAS) 1974; the International Convention on
Standards of Training, Certification and Watchkeeping for Seafarers (STCW) 1978; and the
International Convention on Load Lines 1966.
4
The Canada Shipping Act 2001, Load Line Regulations (SOR/2007-99) and the Safety
Management Regulations (SOR/98-348) have been used to ratify conventions at Canadian
state level. In the United Kingdom (UK) the Merchant Shipping Act 1995, the Merchant
Shipping (Load Line) Regulations 1998 and Merchant Shipping Notices 1812 & 1826
(SOLAS) interpret international codes; while in the United States of America (USA) US Title
33 (Navigation and Navigable Waters) and Title 46 (Shipping) are examples of ratifying
legislation.
While this legislative framework can benefit safety of harbour towage operations, tugs can
also fall below Gross Tonnage thresholds for many international conventions. Although
companies investigated by the author voluntarily complied with SOLAS Chapter IX
(Management for the Safe Operation of Ships) there is no legal requirement for vessels
under five hundred Gross Tonnes to operate Safety Management Systems.
Equally, in the three states examined the Load Line Regulations applied to vessels of 150
gross tonnes (GT) or more, and 24 m or more in length; this may create exceptions for
certain vessels engaged on sheltered waters voyages, and exclude particular categories of
tug.
While the Maritime Labour Convention 2006 (ILO, 2013) makes provision for suitable hours
of rest for mariners, in the United Kingdom, ‘MSN 1767 Hours of Work, Safe Manning and
Watchkeeping Regulations, do not apply to ‘seafarers engaged on tugs in categorised
waters’.
Ratification of International legislation at state level, can also create opportunity for variation
in interpretation. The UK Boatmasters’ Regulations [SI 2006 No. 3223] creates a licensing
system for tugmasters of vessels falling outside conventional legislation (under Workboat
Codes). The UK Port Marine Safety Code facilitates ports to develop management systems,
supporting safe tug operations: the port of Heysham (2011) creating the ‘Minimum safety
standards for tug boats operating within the port’s jurisdiction’. However this legislative
programme is not universal, to all states engaged with harbour towage operations.
Positive progress was witnessed in non legislative spheres, with improved understanding of
technical aspects. Henson (2012) points out that proper tow planning can improve safety of
operations; proposing that the use of ‘Tractor tugs, and tugs with propulsion units forward,
are much safer to operate as bow tugs, as they can better compensate for the interactions
forces’.
5
In a combined report with Merkelbach and van Wijnen (2013) following a comprehensive
international survey, they highlight the importance of maintaining ‘a safe speed’, with ‘all
parties following correct and safe procedures, when making a towline connection’.
New tug designs such as the EDDY are being developed, with one thruster forward and one
aft in order to improve handling; while new towing systems are in operation, such as the
Rotor, which helps minimise towline friction.
In addition to safe speed and safe procedures Henson, Merkelbach and van Wijnen (2013)
identify the importance of ‘comprehensive training underpinned by experience, for tug
masters, pilots and ship’s captains, ensuring optimum team working between all those
involved in safe harbour towage operations’.
6
The object of this research was to identify and quantify independent variables causing a
threat to harbour towage safety, the dependent variable.
The difficulty of sampling a population not involved in safety incidents, in order to establish a
control group, raised investigative problems. Experimental research would be an ideal
methodology; however, it would not be practical or ethical to have accidents under controlled
conditions, in order to determine Risk Factors.
Analytical survey, through exploration of associations between variables, was therefore
considered a more appropriate choice; however, since a population studied may represent a
particular segment, the methodology would need to allow for statistically skewed
distributions.
As an exploratory non experimental project, the research was not able to follow a highly
structured deductive approach and control variables to generate data for analysis. The
project therefore also relied upon a Phenomenological approach, gathering contextual
descriptions of people’s experiences, and as an open ended enquiry using active
experience, it contained Heuristic elements.
The survey, sampling over a discrete 3 month period was cross-sectional; however because
it also relied upon data collected over a period of ten years it incorporated longitudinal
qualities.
The project used an Interpretivist perspective, employing Multiple Methods to triangulate
results, and was divided into six phases (See table 1.) to allow analysis and production of a
report by June 2013.
The three research methods used were:
1. Grounded Theory (Marshall, 1996) qualitative Interview and observational analysis of
expert witness opinion;
2. Quantitative sampling analysis of existing accident Case Study data;
3. Questionnaire survey of practitioners’ professional experience.
2.0 Methodology
7
Phase Activity Completion date
One Production of Research Proposal 26/10/2012
Two Planning of the project 16/11/2012
Three Research literature review 28/11/2012
Four Collection of primary and secondary data 28/01/2013
Five Analysis and interpretation of data 30/03/2013
Six Production of research report 31/05/2013
Table 1: Research Project Programme.
The first stage involved interviews of experts. Data was then analysed using Grounded
Theory coded analysis (Calman, 2011) to provide a depth of perspective.
The second stage consisted of an analysis of secondary data (University of Southampton,
2012) followed by statistical testing, to establish any correlation. The use of ninety case
studies (thirty each from three separate states) aimed to reduce sampling error.
The third stage consisted of a questionnaire survey of current practitioners. This used a
Likert style questionnaire (Social Research Methods, 2006) to enable comparison of
independent variables and to cross check results. Its purpose was to provide specific
contemporary figures, to help identify patterns of safety incident type, cause, result,
frequency and criticality.
During the process, safety Risk Factors were identified, critically evaluated and categorised.
Their likelihood and severity were measured, and this information was used to test
hypotheses.
Since each technique was different, some adjustment was necessary to enable comparison
of the three separate samples and allow triangulation (Holtzhausen, 2001) to help validate
conclusions.
8
2.1 Secondary data
2.1.1 General
The project employed two stages of secondary data collection. The first stage was a
preliminary exploration of the issues, to ensure a full range of incidents and factors were
investigated. The second stage involved collection of a portfolio of Case Studies from the
databases of three flag states.
2.1.2 Textbooks, journals and articles
A mixture of textbooks, journals and articles were investigated. This included Maritime
Safety Agency, tug company, harbour authority and other organisation safety incident
reports (MAIB, 2012; DSB, 2011). Manuals and professional books provided specialist
technical advice on safety risk factors and best practice guidance (Slesinger, 2010;
Livingstone, 2006). Trade and industry journals afforded additional expert opinion and
contextual information (International tug and OSV Magazine, 2013).
2.1.3 Internet
Internet search-engines, databases, and news agencies were used to access current
information (Intute, 2006). The Internet was also used to investigate university library,
government and company databases (KMSB, 2012; EMSA, 2012).
In addition, the internet provided a means to contact organisations & individuals, and to
dispatch documents (International Tug masters Association, 2012; UK Harbour Masters
Association).
9
2.2 Case Studies
2.2.1 General
It was considered impracticable to survey every state in the time scale available; three states
were therefore selected from a shortlist thirty four, using criteria including:
possession of a maritime border and an established port system;
provision of a readily accessible maritime administrative system;
has globally representative characteristics;
translates safety reports in to English.
The final selection, providing representation of harbour towage operations accidents, were:
Canada;
United Kingdom;
United States of America.
2.2.2 Development of the Harbour Towage Safety Risk: Excel
Spreadsheet Proforma
An Excel spreadsheet template was developed to gather data on risks to safety in harbour
towage operations; three example maritime safety agency incident case studies and a
United Kingdom Marine Accident Investigation Bureau accident data sheet, provided a
model of potential factors.
2.2.3 Populating the Excel Spreadsheet
Detailed data on risks to safety were collected from thirty most recent and available accident
case studies, in three separate states. This information was critically analysed, to ensure
data quality and validity:
only Case Studies produced by government safety agencies were accepted;
incident dates and names were cross checked to prevent duplication.
10
The following generic data was gathered from each of the Case Studies and entered on to
the Excel Spreadsheet:
General incident and vessel particulars (date, vessel name, Gross Tonnage & length
overall);
Incident type and consequence (collision or grounding; damage or injury);
Prevailing environmental conditions (weather & sea state);
Towage operation particulars (tug type & tow point);
Findings concerning analysis, causes and conclusions (contributing factors, risks &
potential solutions).
Risk Factors were grouped under generic headings developed from maritime safety agency
reports. Where the lack of a particular factor was reported as a cause, then this would be
recorded; for example, if inadequate safety management systems were identified, this would
be recorded as Risk Factor: ‘Safety Management System’.
Any inapplicable cases studies were identified and removed, where they contained
insufficient verifiable facts.
Remaining case studies were categorised non harbour towage if they were engaged in, for
example:
deep sea towing;
not engaged in towing.
The applicable case studies were then statistically analysed, compared and triangulated with
the other surveys.
11
2.3 Primary Data: Questionnaire
2.3.1 General
A questionnaire approach was selected to gather primary data because it:
provided a systematic quantitative measure, describing, comparing and explaining
contemporary factors effecting harbour towage operations safety (Sapsford, 1999);
enabled gathering large volumes of information over a short time period, from across
the globe.
The questionnaire process was planned to ensure systematic data collection; efforts were
made to standardise the process and eliminate error, by following set procedures and
keeping robust records.
The questionnaire was divided into five sections (See Table 2.). Most questions were
closed, dealing with factual, measureable information; although there were opportunities to
provide additional alternatives or descriptive facts.
Section five used a unipolar, Likert, forced choice, response scale, to grade degree of
applicability. This was chosen to reduce selection of a middle ‘neutral’ option, and to
motivate greater consideration of all explanations.
Section Content
One Instructions and further information concerning the research project.
Two Factual details concerning the particular harbour towage operation.
Three Environmental factors affecting the operation.
Four Details of the risk or safety issue encountered.
Five Risk Factors considered to be causing the risk or safety issue.
Table 2: Questionnaire Sections.
12
2.3.2 Process
A pre-test survey was conducted, by providing the draft questionnaire to two non-
participants, to check for comprehension, construction and clarity. Research ethics for the
whole project, were addressed using the Solent Research Ethics Release (2013) process.
To maximise participation, a publicity plan was produced (See Table 3.). This enabled
identification of potential sources and methods to advertise the project, together with
timetabling of target dates. Any publicity through conventional media, with its long lead-time,
necessitated prompt production of promotional material. The Internet and electronic
communications media were however key to ensuring global participation, in the short time
frame available.
Target group Method of publicising Deadline
Harbour Towage
Organisations
Direct email, provision of electronic questionnaire, and
follow-up correspondence to relevant organisations.
10/11/2012
Professional
Mariners
Contact Maritime Media Companies, with a press
release and follow up correspondence.
10/11/2012
Pilotage
Organisations
Direct email, provision of electronic questionnaire, and
follow-up correspondence.
20/11/2012
Harbour Towage
Companies
Direct email, provision of electronic questionnaire, and
follow-up correspondence.
20/11/2012
Other Web-site, for provision of research project information
and questionnaires.
30/11/2012
Reactive Contacts Direct provision of questionnaire. 28/02/2013
Table 3: Publicity Programme.
To properly organise administration, with its reliance upon electronic communications,
setting up of a dedicated e-mail facility was required: [email protected]
Creation of a “Tug Safety” research project web-site helped authenticate its provenance and
supplied information to participants: http://mahara.solent.ac.uk/view/view.php?id=66091
13
2.3.3 Target Participation
In consultation with the Solent University Research Project Co-ordinator, a target of 30
questionnaires was decided, as valid and practicable. To achieve this, the questionnaire
was distributed to a stratified group of one hundred and thirty five interested organisations
and individuals (See table 4.).
Group Total questionnaires distributed
Class surveyors 5
Harbourmasters 20
Maritime Administrators 5
Marine Insurers 5
Maritime legislators 5
Pilots 20
Ship’s crews 20
Towage Company Managers 10
Tug crews 30
Tug interest organisations 10
Tug shipbuilders 5
TOTAL 135
Table 4: Questionnaire Distribution Groups
All participants were volunteers, who were informed of the purpose of the project and no
pressure was applied to participate. To ensure privacy, the researcher was the only person
to contact participants, and submissions were entered on to a single computer, to which only
the author had password access. No copies of submissions were made, no details were
released to third parties, and all submissions are to be destroyed on 30th July 2013.
There were three main means of participation:
participants could receive a questionnaire forwarded from their employer or
professional body;
a questionnaire could be sent electronically from the ‘Tug.Safety’ e-mail account;
a questionnaire could be downloaded from the Tug Safety web-site.
14
2.3.4 Quality Control
While without password protection there was less control over respondents, potentially
reducing data validity, this had to be balanced against the practicalities of maintaining a
global response. To mitigate this, contact details of participants were kept for subsequent
validation, should this prove necessary.
Response reliability may have been increased, and socially desirable response bias
reduced, through the anonymity of online surveying. Use of a global population may have
reduced sampling error, diversifying the population and randomly distributing any errors; this
was particularly important as volunteers, rather than a random sample, might increase
potential for skewed distributions.
Questionnaires were self administered, using instructions contained on the questionnaire
and returned directly to the dedicated e-mail account for analysis (See Annex A.). On
receipt of completed questionnaires, an acknowledgement was sent, together with details of
how follow-up information could be obtained.
2.3.5 Analysis
Each Questionnaire was assigned a unique reference number and the data was cleaned
(checking for obvious errors & ineligibility). The data was then collated and coded for
quantitative analysis on an Excel spreadsheet.
Each Likert item was treated as ordinal data and analyzed separately; when using only four
significance levels, it could not be assumed that respondents perceive the difference
between adjacent levels, as equidistant. If treated as ordinal data, Likert responses could be
analyzed using non-parametric testing.
Data from the Questionnaire survey was combined on the Excel spreadsheet, with that from
the Case Studies. The process used the same format as the Case Studies to enable direct
comparison between surveys.
15
2.4 Grounded Theory
2.4.1 General
A Grounded Theory interview qualitative research process was selected because:
in contrast with the other techniques it was explicitly emergent; it did not test a
hypothesis, but set out to find what theory accounts for a situation (Dick, 2005);
it allowed the study of social interactions & behaviour, measuring attitude & opinion,
as integral factors;
it allowed an in-depth exploration of this relatively new area, where previous research
was limited.
The first stage of the process, involved a literature review. The interview process and
analysis were then planned, to establish systematic sampling and data collection. The
interview process was then pre-tested and all equipment was checked.
A Judgement sample of expert interviewees was targeted: those with over twenty five years
professional experience (a period during which practitioners might reasonably be expected
to have encountered a range of harbour towage operations safety issues). To maximize
variation in experience and avoid subject bias, experts from contrasting operations,
management and regulatory roles were selected, from:
Tug Handlers;
Harbour Pilots;
Harbourmasters;
Maritime Legislators.
All interviewees were volunteers and no pressure was applied to participate. Before
interviews commenced, the process was risk assessed (following Solent University
Guidelines) and interviewees were informed of the purpose of the project.
To ensure confidentiality, only one transcript was produced for each interview, this was
maintained as a controlled document, and no details of interviewees were released to third
parties.
16
2.4.2 Process
Initial contact was made with potential expert interviewees; if they agreed to participate, a
date was decided for subsequent interview. On the agreed date the interview was
conducted, following recommended guidelines (Tellis, 1997).
At this stage, it was not possible to specify a particular sample size, since Grounded Theory
requires repetition, until new data no longer provided new information; a point called
theoretical saturation (Glasser & Strauss, 1967).
Interviews with experts were held, either face to face or over the telephone, and lasted about
45 minutes. The interviews were recorded using audio equipment, allowing the interviewer
to focus on the conversation; recordings were subsequently transcribed for analysis. Back-
up notes were also taken to highlight particular points. Following interview, participants were
thanked and provided with an address to obtain further information.
Interviews were semi-structured, with the participant asked to describe and reflect upon
experiences of harbour towage safety, using a series of short, clear prompt questions,
concerning tug operations and potential threats to safety, where necessary. The participant
was active while the interviewer listened actively.
They commenced with ‘Open-ended Questions’ (Charmaz, 2006) concerning the experts
background and views on harbour towage safety. ‘Intermediate Questions’ then probed
deeper into safety issues, and ‘Ending Questions’ elicited any concluding remarks.
Interview process and analysis were simultaneous; the first interview provided an initial
question framework, while subsequent interviews evolved iteratively, allowing enquiry to
focus upon apparent patterns (Hoda, 2011). Emerging codes, concepts, and categories
helped structure and systematically capture information from subsequent interviews (Strauss
and Corbin 1994). Written transcripts provided additional observations, further insight and
validation of themes.
During coding, the transcript data was read through several times to get a general
impression and to identify the major ideas, unusual events and deviant cases. It was then
progressively ‘chunked’ into sentences or phrases, to allow ‘Open Coding’ (Hallberg, 2006).
17
Constant comparison was used to ‘Selectively Code’ material into concepts, and to identify
‘Core Categories’ (central themes, reoccurring most frequently, and related to main
categories). The text was systematically marked with the codes or categories, and this
process was repeated several times, until ‘saturation’ was reached. Finally axial coding of
the transcripts was used to compare, identify and verify connections or relationships
between categories and concepts.
Analysis and coding was completed before other data sets were examined, in order to keep
an open mind and so reduce preconceptions. Once the interviews had been coded and
categorised the resulting Grounded Theory was triangulated with the other research
techniques.
18
3.1 Case Study Account
A total of ninety Case Studies (CS) were selected; thirty each from three states. Five CS
were inadmissible through insufficient validation (See Table 5.). Of the eighty five
admissible CS, fifty eight were classified as Harbour Towage (HT) and the remaining twenty
seven Non Harbour Towage (NHT).
State Inadmissible
Case Studies
Harbour
Towage
Non Harbour
Towage
Total
Canada 1 20 9 30
United Kingdom 0 20 10 30
United States of
America
4 18 8 30
Total 5 58 27 90
Table 5: Distribution of Harbour Towage and Non Harbour Towage Case Studies
Ninety five percent of the CS were classed as Accidents; of the remaining CS one was
classed an Incident, one a Near Miss and two as Other2. Thirty nine of the CS involved
Collision, twenty four Capsize and twenty Grounding (See Graph 1.).
Graph 1: Apportionment of Accident Type (number)
2. Figures may exceed total Case Studies, since one event may lead to several consequences.
39
20
24
3
6 Collision
Grounding
Capize / Founder
Fire
Other
3.0 Results
19
Analysis of incident consequence indicates that seventy two resulted in Damage, thirty three
in Injury and twenty in Loss of Life (See Graph 2.).
Graph 2: Apportionment of Event Consequence (number)
Analysis of the towage operation indicated, fifty four percent concerned Conventional tugs,
while thirty five percent were an undetermined tug type (See Graph 3.).
Graph 3: Distribution of Tug Type (percentage)
Forty eight percent of tugs were Moderately powered (See Table 7) while twenty three
percent were Medium powered or were Unspecified (See Table 4.).
Graph 4: Distribution of Tug Bollard Pull (percentage)
20
33
9
72
Loss of Life
Injury
Pollution
Damage
54.1
4.7
5.9
35.3
Conventional
ASD
VS
Unknown
48.2
23.6
4.7
23.5
Moderate (<30t)
Medium (31-65t)
High (>65t)
Unspecified
20
Forty four percent of events involved Towing from Forward, twenty percent Pushing, while
twenty two percent were unspecified (See Graph 5.).
Graph 5: Distribution of Tow Position (percentage)
The majority of events (forty six) concerned barges (See Graph 6.).
Graph 6: Towed Vessel Type (number)
The majority of towed vessels (sixty seven percent) had broad bow forms (See Graph 7.).
Graph 7: Bow Form Distribution
44.7
8.2
20
22.4
4.7 Tow fwd
To aft
Push
Other / unknown
Amidships
12
9
46
18
Tanker, Bulk Carrier etc.
Container, RoRo, General Cargo, etc.
Barge
Unspecified
10.6
67.1
22.3
Fine to Moderate
Broad
Unspecified
21
The majority of towed vessels (fifty percent) were classed Small (under 10,000 tonnes
deadweight) while eleven percent were Handy (See Table 8) or were Large (MAN, 2007).
There were no Very Large vessels (over 160,000 tonnes deadweight) while twenty five
percent were of unspecified size (See Graph 8.).
Graph 8: Distribution of Towed Vessels Size (Deadweight Category: percentage)
50.6
11.8
11.8
0
25.8
<10K t
10-50K t
51-160K t
>160K t
Unspecified
22
3.2 Questionnaire Account
Thirty two questionnaires were received by email; nineteen of these were submitted by Tug
Masters, nine by Pilots and four by Ship’s Masters. Submissions were received from
fourteen countries (See Table 6.) and all were considered valid.
Table 6: Questionnaire source
Analysis of the Mean, Median, Mode and Standard Deviation indicated that the data was not
Normally Distributed, with a sample Risk Factor frequency histogram plot indicating a
positive skew.
State Number of
questionnaires
Australia 3
Belgium 1
Chile 1
Finland 1
Ireland 1
Italy 4
Latvia 1
Netherlands 2
New Zealand 2
Poland 1
Portugal 1
Singapore 1
United Kingdom 7
United States 1
Unknown 5
TOTAL 32
23
Sixteen of the Questionnaires concerned Near Misses, there were seven Incidents, five
Challenging operations, three Accidents and one questionnaire was categorised Other (See
Graph 9.).
Graph 9: Safety Occurrence Description (number)
Questionnaire potential for Collision was eighty seven percent, Grounding thirty seven
percent and Capsize or Foundering thirty four percent3.
Graph 10: Safety Occurrence Potential Result (percentage)
3. Figures total over one hundred percent, since a single safety occurrence can lead to multiple results.
5
16
7
3 1
Challenging
Near Miss
Incident
Accident
Other
87.5
37.5 34.8
0 3.1
0 20 40 60 80
100
Pe
rce
nta
ge P
ote
nti
al
Event
24
A breakdown of consequences from a safety occurrence includes ninety three percent
potential for Damage and sixty eight percent potential for Injury. There was fifty percent
potential for Loss of Life, with sixty five percent potential for Pollution (See Graph 11.).
Graph 11: Safety Occurrence Potential Consequence (Average Likelihood)
Eighteen of the Tugs described were Azimuth Stern Drive (ASD) eight were Conventional,
five had Voith Schneider propulsion systems and one was Unspecified (See Graph 12.).
Graph 12: Distribution of Tug Type (number)
Twenty eight percent of tugs were Moderate (See Table 7.) fifty nine percent Medium and
nine percent were High powered (See Graph 13). All of the Conventional tugs were
Moderate powered, while all except two of the ASD tugs were Medium powered.
Graph 13: Distribution of Tug Power Table 7: Tug Power Categories (percentage)
50
68.8 65.6
93.8
0 20 40 60 80
100
Loss of Life
Injury Pollution Damage
Po
ten
tial
(p
erc
en
tage
)
Consequence
8
18
5
1 Conventional
ASD
VS
Unspecified
28.1
59.4
9.4 3.1
Moderate (<30t)
Medium (31-65t)
High (>65t)
Unspecified
Tug Power Bollard
Pull (BP):
tonnes
Moderate < 30
Medium 31 - 65
High >65
25
Ninety four percent of safety occurrences involved use of a line (either Push/Pull or Towing
on a Line); ninety one percent of cases were using the Tug’s Line (See Graph 14.).
Graph 14: Tow Operation Type Distribution
Thirty seven percent of vessels were categorised Container, RoRo or General Cargo, twenty
five percent were Tankers, Gas or Bulk Carriers, three percent were Barges and the
remainder were Unspecified (See Graph 15.).
Graph 15: Towed Vessel Category Distribution
56.2 37.5
6.3
0 20 40 60
Pe
rce
nta
ge
Tow Type
34.4
3.1
25
37.5
0
5
10
15
20
25
30
35
40
Fre
qu
en
cy (
pe
rce
nta
ge)
Vessel Type
26
Thirty four percent of vessels had fine formed bows, twenty one percent moderate and
twenty eight percent were broad bowed (See Graph 16.).
Graph 16: Towed Vessel Bow Form Distribution
Twelve percent of Towed Vessels were Small (under ten thousand tonnes Deadweight) and
Very Large (See Table 8) thirty one percent were Large, twenty eight percent Handy, and
sixteen percent were of Unspecified size (See Graph 17 and Table 8.).
Graph 17: Towed Vessel size Category Table 8: Distribution of Towed
Vessel Deadweight Categories
(MAN, 2007)
15.6
34.4
21.9
28.1
0
5
10
15
20
25
30
35
40
Fre
qu
en
cy (
pe
rce
nta
ge)
Bow Form
15.6 12.5
28.1
31.3
12.5
0
5
10
15
20
25
30
35
Fre
qu
en
cy (
pe
rce
nta
ge)
Vessel Size (thousand tonnes Deadweight)
Towed
Vessel
Size
Category
Deadweight
(tonnes)
Small <10,000
Handy 10,001 – 50,000
Large 50,001 –
160,000
Very Large > 160,000
27
Environmental conditions varied; Modal wind states were Moderate (between Beaufort Wind
Force four and six). Modal swell conditions were Calm (under 0.2m swell height); although
they were categorised Rough (1 to 1.5m) on six percent of occasions and Heavy (over 1.5m)
on nine percent of occasions. Modal current conditions were Low (less than 1 knot) with
Moderate current on nineteen percent and Strong current on nine percent of occasions. Fog
was present on nine percent of occasions (See Graph 18.).
Graph 18: Prevailing Environmental Conditions
0
1
2
3
4
5
6
7
8
9
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
Me
asu
rem
en
t
Questionnaire No.
Wind Strength (Beaufort Force) Swell Height (metres) Current Strength (Knots)
28
3.3 Expert Interview Account
Five Experts were interviewed. Interviewees had between twenty seven and fifty two years
employment experience in maritime operations; roles performed included as Tug Master,
Ship Master, Harbour Master, Pilot, Superintendent, Regulatory Surveyor, Class Surveyor
and Marine Consultant.
Speed, Interaction and Girting: Experts identified vessel speed as a critical risk factor,
(and associated with this, the effect of interaction); the faster the vessels, the less safe the
operation, and the increased potential for interaction or girting.
Code Sample statement
Speed &
Interaction
“speed is a big factor, due to interaction, it is a major cause…”
Girting “[girting] still cont. to happen, even though there are ways of reducing the
probability, it continues to occur…”
Table 9: Evidence Category: Speed, Interaction & Girting
Extreme Conditions: Experts identified extreme environmental conditions as a
critical risk factor; in particular physical limitations posed by proximity to vessels & structures,
and the effects of swell, wind, current & fog.
Code Sample statement
Swell “every towline breaking with me , it’s been swell conditions, snatching …”
Proximity “thrusters are a problem [making] it difficult to maintain position & stop
yourself from being washed away…”
Wind /
current
“… it was not long before I realized that the four tugs could not hold the
vessel, and that we were being blown to ….”
Table 10: Evidence Category: Extreme Conditions
Legislation: Lack of regulatory oversight was identified as a particular risk to
safety; with potential for unclassified vessel’s to go unmonitored.
Code Sample statement
Legislative
equitability
“risk assessment, safety management, planned maintenance; all these
things have been brought in for big ships, but maybe the small ones have
slipped through under the wire …”
Monitoring “anything less than 12 pass and anything less than 24m [gets ignored]…”
Table 11: Evidence Category: Legislation
29
Tow Planning & Command: A lack of planning and the command of towage
operations were identified as risks to safety.
Code Sample statement
Planning “they’d planned the job up until [buoy x], but not for the rest of the pilotage
…”
Command “there were really two people in charge of the operation…”
Table 12: Evidence Category: Tow Planning & Command
Maintenance: A lack of maintenance (in particular of critical / safety equipment) was
identified.
Code Sample statement
Maintenance
Failures
“you’ve got the classic tow hook, not being maintained, it has been the
cause of many accidents …”
Table 13: Evidence Category: Maintenance
Design & Complexity: Advances in vessel design and increasing complexity were
seen as risk factors.
Code Sample statement
Design “assisted vessels, become ever bigger, faster, heavier, deeper, [with
operations] all at more speed …”
Complexity “… they’ve got so sophisticated nowadays, and it’s the engine that does you;
it stops, because some silly alarm goes off …”
Table 14: Evidence Category: Design & Complexity
Seamanship & Rope Management: Seamanship & Rope Management were
identified as risks factors.
Code Sample statement
Rope
Management
“… a man gets hand trapped in towing hawser …”
Table 15: Evidence Category: Seamanship & Rope Management
30
Crewing: Lack of sufficient crew of adequate professional training were
identified as risk factors.
Code Sample statement
3 Man
Crewing
“In big port entrances the lookout is left very much to other vessels ….”
Table 16: Evidence Category: Crewing
Communications: Poor communications were only explicitly raised by one expert,
although associated ‘communications’ factors were cited.
Stability: Lack of stability was identified as a high severity risk to safety; in
particular this concerned loss of watertight integrity resulting from insecure openings.
Code Sample statement
Watertight
Integrity
“tugs have got such bloody good stability that you can yank them right over
and they will bounce back. But they won’t bounce back if you’ve got a door
open ...”
“it’s always someone leaving the door open …”
Table 17: Evidence Category: Stability
Human Factors: Human Factors were cited as a risk to safety; but there were
contradicting statements advocating that the situation had improved.
Code Sample statement
Fatigue #1 “the crew were all seasick, and the master couldn’t physically do anymore
…”; versus:
Fatigue #2 “in the old days there was less control, you had to keep working …”
Table 18: Evidence Category: Fatigue
Training: The need for improved Training was cited as a risk factor. This
element related to other Risk Factors including: Following Operational Procedures, Tow
Planning, Tug Handling and Communications; additional linked codings’ included personal
qualities, and the importance of team working & judgement.
Code Sample statements
Training “a Voith Schneider captain is not an ASD tug captain and vice-versa …”
“lack of thought by [the assisted] vessel’s bridge team was a problem…”
Table 19: Evidence Category: Training
31
Personal Qualities & Negative Attitudes: Related to Training, the importance of Personal
Qualities & Judgement, were identified as safety factors.
Code Sample statement
Personal
Qualities &
Judgement,
“you need to be experienced enough to say yes I can do it, or no it’s too
risky - it’s a fine line…”
“it needs a certain personality able enough to cope with the wildness [power]
of this tug, in fact it must be an anti macho figure…”
Negative
attitudes
“one [problem] is complacency; old skippers saying, I’ve always done it this
way…”
“there is no reason for this to occur other than negligence …”
Table 20: Evidence Category: Personal Qualities & Negative Attitudes
Time: Lack of Time was identified as a threat to safety of operations; both internal
(e.g. as a human quality of rushing) and external (e.g. commercial pressure).
Code Sample statements
Internal “people, to gain time for whatever reason, end up in all sorts of risks…”
External “to be able to learn to drive safely, skippers need to be given time …”
Table 21: Evidence Category: Time
32
3.4 Comparison of Harbour Towage and Non Harbour
Towage Data
Harbour Towage (HT) operations had eighteen percent more Collisions, while Non Harbour
Towage (NHT) operations had thirty six percent more Groundings (See Graph 19).
Graph 19: HT to NHT Incident Category Comparison (relative frequency)
Harbour towage had at least twice the frequency in all Consequence categories (See Graph
20).
Graph 20: HT to NHT Comparison of Consequences
33.3
48.1
25.9
11.1 7.4
51.7
12.1
29.3
0
6.9
0
10
20
30
40
50
60
Collision Grounding Capize / Founder
Fire Other
Re
lati
ve F
req
ue
ncy
Event
Non Harbour Towage Harbour Towage
5
8 3
20
15
25
6
52
0
10
20
30
40
50
60
Loss of Life Injury Pollution Damage
Fre
qu
en
cy
Consequnce
Non Harbour Towage Harbour Towage
33
Forty percent of NHT operations involved towing from Forward, compared to forty six
percent of HT (See Graph 21).
Graph 21: Comparison of HT and NHT Towage Position (relative frequency)
Comparison between HT and NHT incidents, indicated similar proportions of barges (fifty five
to fifty three percent); however there was noticeable variation in all other categories (See
Graph 22).
Graph 22: Comparison of Towed Vessel Type between HT and NHT Case Studies
40.7
0
7.4
40.8
11.1
46.5
12.1
25.9
13.8
1.7
0
5
10
15
20
25
30
35
40
45
50
Tow fwd To aft Push Other / unknown
Amidships
Re
lati
ve F
req
ue
rncy
Towing position
Non Harbour Towage Harbour Towage
0 0
55.6
44.4
20.7 15.5
53.5
10.3
0
10
20
30
40
50
60
Tanker, Bulk Carrier etc.
Container, RoRo, General
Cargo, etc.
Barge Other
Pe
rce
nta
ge
Vessel Type Non Harbour Towage Harbour Towage
34
Fifty one percent of NHT Towed Vessels were Broad Bowed, compared to seventy four
percent of HT.
Graph 23: Comparison between HT and NHT Bow Form Distribution
Comparison between HT and NHT incidents, indicated similar proportions of Small Towed
Vessels (under ten thousand tonnes Deadweight): fifty and fifty one percent respectively;
however there was noticeable variation amongst all other categories (See Graph 24).
Graph 24: Comparison of Towed Vessel Size
0
51.9 48.2
15.5
74.1
10.4
0
20
40
60
80
Fine to Moderate
Broad Unspecified
Pe
rce
nta
ge
Bow Form Non Harbour Towage Harbour Towage
48.1 51.9
0 0 0
15.6
50
17.2 17.2
0
0
10
20
30
40
50
60
Unknown <10K t 10-50K t 51-160K t >160K t
Pe
rce
nta
ge
Vessel Deadweight Category Non Harbour Towage Harbour Towage
35
Graph 25: Case Study Comparison of HT and NHT Risk Factor Frequencies
Risk Factor comparison between harbour and non harbour towage produced several
findings (See Graph 25). Eleven Risk Factors were present only in harbour towage
operations (in rank frequency):
Tug Handling (fifty three percent);
Rope Management & Seamanship (thirty four percent);
Interaction (thirty two percent);
Girting (twenty five percent);
Current (twenty four percent);
Ship Securing Arrangements (seventeen percent);
Ship Size [towed] (seventeen percent);
Personal Protective Equipment (PPE) (thirteen percent);
Tug Type (thirteen percent);
Communications Equipment (eight percent);
General Purpose Manning (six percent).
0
0
11
.1
66
.7
0
0
0
3.7
0
0
0
7.4
11
.1
0
11
.1 7.4
0
29
.6
0
25
.9
74
.1
51
.9
77
.8
11
.1
11
.1
74
.1
51
.9
0
18
.5
32
.8 25
.9
75
.9
44
.8
53
.5
34
.5
6.9
27
.6
17
.2
0
13
.8
63
.8
5.2
24
.1
8.6
10
.4
17
.2
25
.9
8.6
43
.1
82
.8
56
.9
87
.9
10
.4
17
.2
41
.4
48
.3
13
.8
27
.6
0
10
20
30
40
50
60
70
80
90
100
Inte
ract
ion
Gir
tin
g
Tow
Pla
nn
ing
Pas
sage
Pla
nn
ing
Tug
Han
dlin
g
Ro
pe
/ Se
aman
ship
Gen
eral
Pu
rpo
se …
Spee
d
Ship
Siz
e
Ship
Po
wer
Tug
Typ
e
Man
ou
evri
ng
Spac
e
Swel
l
Cu
rren
t
Win
d
Vis
ibili
ty
Ship
Sec
uri
ng …
Tug
Equ
ipm
ent
Co
mm
un
icat
ion
…
Co
mm
un
icat
ion
s
Hu
man
Fac
tors
Trai
nin
g
Man
agem
ent
Syst
ems
Cre
w N
um
ber
s
Bri
dge
/ E
qu
ipm
ent …
Wat
chke
epin
g
Follo
win
g O
per
atio
nal
…
PP
E
Oth
er
Fre
qu
en
cy (
pe
rce
nta
ge)
Risk Factor
Non Harbour Towage Harbour Towage
36
Three Risk Factors were noticeably more frequent in Harbour Towage Operations:
Tow Planning (seventy five percent);
Manoeuvring Space (sixty three percent);
Speed (twenty seven percent).
Six Risk Factors had prominent frequencies in HT (and NHT) operations (in rank average
frequency):
Management Systems (eighty two percent);
Human Factors (seventy eight percent);
Watchkeeping (fifty seven percent);
Passage Planning (fifty five percent);
Training (fifty four percent);
Following Operational Procedures (fifty percent);
In addition, the average total number of risk factors was greater for HT (nine) than for NHT
operations (five).
Graph 26: Comparison of HT and NHT Risk Factor Volumes
0
2
4
6
8
10
12
14
16
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27
Nu
mb
er
of
app
licab
le R
isk
Fact
ors
(to
tal)
Case Study (averaged)
Harbour Towage Occurences Non Harbour Towage Occurences
37
3.5 Hypothesis Testing: Was there a measureable
difference between Harbour Towage and Non Harbour
Towage Operations (Chi Square test)
A Chi Square test comparing Harbour Towage (HT) and Non Harbour Towage (NHT) Risk
Factors rejected the Null Hypothesis in ten out of ten cases. (The test could not be
performed on other Risk Factors whose Estimated Values were below ten).
Risk Factor Alpha
value
(0.05)
X2
value
Is there a relationship between Risk Factor
and Harbour Towage Operation?
Tow Planning 3.84 38.397 Yes
Passage Planning 3.84 14.356 Yes
Tug Handling 3.84 31 Yes
Rope Management &
Seamanship
3.84 21.861 Yes
Manoeuvring Space 3.84 31.758 Yes
Tug Equipment
Maintenance
3.84 11.420 Yes
Communications 3.84 13.313 Yes
Training 3.84 11.470 Yes
Lookout / Watchkeeping 3.84 18.144 Yes
Following Operational
Procedures
3.84 11.387 Yes
Table 22: Chi Square Test values
The Chi Square test of Risk Factors indicates a detectable difference between harbour
towage and non harbour towage operations (See Table 22.).
38
3.6 Comparison of Quantitative Data
3.6.1 Comparison of Secondary with Primary Harbour Towage
Quantitative Data
The most frequent Event in both Case Studies (fifty eight percent) and Questionnaires
(eighty seven percent) was Collision.
Graph 27: Comparison of CS and QU Events
The most frequent Consequence in both CS (eighty nine percent) and QU (ninety three
percent) was Damage. Both surveys also had a noticeable Loss of Life frequency (thirty
seven percent).
Graph 28: Comparison of Consequence Frequency
58
.6
12
.1
29
.3
0
6.8
87
.5
37
.5
34
.8
0
3.1
73
.05
24
.8
32
.05
0
4.9
5
0
10
20
30
40
50
60
70
80
90
100
Collision Grounding Capize / Founder
Fire Other
Fre
qu
en
cy (
pe
rce
nta
ge)
Event
Case Study Questionnaire Average
25
.9
43
.1
10
.4
89
.7
50
68
.8
65
.6
93
.8
37
.95
55
.95
38
91
.75
0
10
20
30
40
50
60
70
80
90
100
Loss of Life Injury Pollution Damage
Fre
qu
en
cy (
pe
rce
nta
ge)
Consequence Case Study Questionnaire Average
39
Conventional Tugs were most frequent in CS (fifty one percent) whereas ASD Tugs were
most frequent in QU (fifty six percent).
Graph 29: Comparison of Tug Type
Moderately powered Tugs were most frequent in CS (forty six percent) whereas Medium
powered tugs were most frequent in QU (fifty nine percent).
Graph 30: Comparison of Tug Power
51
.7
6.9
6.9
34
.5
25
56
.3
15
.6
3.1
38
.35
31
.6
11
.25
18
.8
0
10
20
30
40
50
60
Conventional ASD VS Unspecified
Fre
qu
en
cy (
pe
rce
nta
ge)
Tug Type
Case Study Questionnaire Average
46
.6
25
.9
34
.5 2
4.1
28
.1
59
.4
9.4
3.1
37
.35
42
.65
21
.95
13
.6
0
10
20
30
40
50
60
70
Moderate (<30t) Medium (31-65t) High (>65t) Unspecified
Fre
qu
en
cy (
pe
rce
nta
ge)
Tug Power
Case Study Questionnaire Average
40
A Forward tow position was most frequent in CS, whereas an Aft tow position was most
common in QU (both forty six percent).
Graph 31: Comparison of Tow Position
Barges were the majority of towed vessels in the CS (fifty three percent) whereas they were
the least frequent Category in the QU (three percent).
Graph 32: Comparison of Towed Vessel Type
46
.6
12
.1
25
.9
0
15
.5
25
46
.9
0
3.1
25
35
.8 29
.5
12
.95
1.5
5
20
.25
0
5
10
15
20
25
30
35
40
45
50
Tow fwd To aft Push Amidships Other / unspecified
Fre
qu
en
cy (
pe
rce
nta
ge)
Tow position
Case Study Questionnaire Average
20
.7
15
.5
53
.5
10
.4
25
37
.5
3.1
34
.4 22
.85
26
.5
28
.3
22
.4
0
10
20
30
40
50
60
Tanker / Bulk carrier
Container / RoRo / General
Cargo
Barge Other / unspecified
Fre
qu
en
cy (
pe
rce
nta
ge)
Towed Vessel Type
Case Study Questionnaire Average
41
The most frequent Bow form in CS was Broad (seventy four percent) whereas a Fine bow
form was most common in QU (thirty four percent).
Graph 33: Towed Vessel Bow Form Comparison
The most frequent Towed Vessel category in the CS was Small (fifty percent) (See Table 8)
whereas the most frequent category in QU was Large (thirty one percent).
Graph 34: Towed Vessel Size Comparison
0
15
.5
74
.1
10
.4
34
.4
21
.9
28
.1
12
.5
17
.2
18
.7
51
.1
11
.45
0
10
20
30
40
50
60
70
80
Fine Moderate Broad Unspecified
Fre
qu
en
cy (
pe
rce
nta
ge)
Bow Form Case Study Questionnaire Average
50
17
.2
17
.2
0
15
.5
12
.5
28
.1
31
.3
12
.5
15
.6
31
.25
22
.65
24
.25
6.2
5
15
.55
0
10
20
30
40
50
60
<10K t 10-50K t 51-160K t >160K t Unspecified
Fre
qu
en
cy (
pe
rce
nta
ge)
Towed Vessel Deadweight (tonnes)
Case Study Questionnaire Average
42
The most frequent Risk Factors in both the Case Study and Questionnaire were (average):
Human Factors (seventy five percent);
Tow Planning (seventy three percent);
Manoeuvring Space (sixty nine percent);
Training (sixty two percent);
Tug Handling (fifty six percent).
Graph 35: Comparison of Questionnaire & Case Study Harbour Towage Risk Factors
Other notable Risk Factors in both surveys were (average):
Communications (forty one percent);
Interaction (thirty nine percent);
Tug Equipment / Maintenance (thirty six percent);
Girting (thirty three percent).
Ship Speed and to a lesser extent Size also had notable frequencies (See Graph 36);
however there was clear variation between the CS and QU data:
Ship Speed (average forty eight percent, with forty one percent variation);
Ship Size (average thirty two percent, with twenty seven percent variation).
32
.8 25
.9
75
.9
53
.4
27
.6
17
.2
0
13
.8
63
.8
5.2
24
.1
8.6
10
.3
17
.2
25
.9
8.6
43
.1
82
.8
56
.9 4
6.9
40
.6
71
.9
59
.4
68
.8
46
.9
46
.9
31
.3
75
40
.6
28
.1
37
.5
9.4
40
.6
46
.9
18
.8
40
.6
68
.8
68
.8
39
.85
33
.25
73
.9
56
.4 4
8.2
32
.05
23
.45
22
.55
69
.4
22
.9
26
.1
23
.05
9.8
5
28
.9
36
.4
13
.7
41
.85
75
.8
62
.85
0
10
20
30
40
50
60
70
80
90
Fre
qu
en
cy (
pe
rce
nta
ge)
Risk Factor
Case Study Harbour Towage Data Questionnaire Harbour Towage Data Average Harbour Towage Data
43
With the exception of Visibility, the remaining Risk Factors were presents on average, in
over ten percent of events.
Graph 36: Rank Difference Case Study Versus Questionnaire
The largest rank movement resulting from the different frequencies between the two surveys
was Ship Power (eleven places); Current, Swell, Communications, Girting and Visibility also
experienced rank movements of between four and six places (See Graph 37.).
-8
-6
-4
-2
0
2
4
6
8
10
12
Viz
Co
mm
un
icat
ion
…
Cu
rren
t
Tug
Typ
e
Win
d
Gir
tin
g
Swel
l
Secu
rin
g A
rran
gem
ents
Co
mm
un
icat
ion
s
Inte
ract
ion
Ship
Siz
e
Ship
Po
wer
Tug
equ
ipm
ent
Tug
Han
dlin
g
Ship
Sp
eed
Hu
man
fac
tors
Trai
nin
g
Tow
pla
nn
ing
Man
ou
evri
ng
Spac
e
44
3.6.2 Additional Risk Factors
Eight Risk Factors were identified in the Case Studies (CS) and the Expert Interviews (EI)
but were not included in the Questionnaire (QU); four of these had frequencies of over forty
percent (See Graph 38):
Safety Management Systems (eighty seven percent);
Following Operational Procedures (forty eight percent);
Passage Planning (forty four percent);
Watchkeeping (forty one percent).
Graph 37: Additional Risk Factors identified in Case Studies
Stability, Time, and Crew Qualities & Attitudes were identified in the Expert Interviews, but
were not explicitly identified in the Case Studies or Questionnaires.
44
.8 34
.5
6.9
87
.9
10
.4
17
.2
41
.4
48
.3
0 10 20 30 40 50 60 70 80 90
100
Fre
qu
en
cy (
pe
rce
nta
ge)
Risk Factor
45
3.7 Risk Factors
3.7.1 Questionnaire Risk Factor Frequency
According to the Questionnaires (QU) the most frequently occurring risk factors, present in
over half of events (in rank order were):
Manoeuvring Space;
Tow Planning;
Speed;
Human Factors;
Training;
Tug Handling;
Graph 38: Comparison of Weighted and Un-weighted Questionnaire Risk Factors
15
13
23
19
22
15
15
10
24
13
9
12
3
13
15
6
13
22
22
29
20
42
37
51
22
32
17
44
19
13
18
5
22
27
12
22
41
37
0
10
20
30
40
50
60
Safe
ty F
acto
r Fr
eq
ue
ncy
SafetyFactor
Unweighted Weighted (by perceived importance)
46
3.7.2 Questionnaire Risk Factor Weighting
Questionnaire Risk Factor significance weighted by perceived importance, reflected and
amplified Risk Factor frequency (See Graph 40.); the most notable amplification being for:
Speed (one hundred and thirty one percent);
Ship Power [towed] (one hundred and thirteen percent);
Communications Equipment (one hundred percent);
Interaction (ninety three percent);
Tug Handling (ninety four percent);
Graph 39: Risk Factor Percentage increase due to perceived importance
93
.3
53
.9
82
.6
94
.7
13
1.8
46
.7
11
3.3
70
83
.3
46
.2
44
.4
50
66
.7
69
.2
80
10
0
69
.2
86
.4
68
.2
0
20
40
60
80
100
120
140
pe
rce
nta
ge in
cre
ase
Risk Factor
47
3.7.3 Questionnaire Risk Factor alteration in ranking
This altered the Risk Factor ranking in ten cases; the most notable movement being for
Speed (increase of four positions) and Ship Size (decrease of three positions) (See Graph
41).
Graph 40: Risk Factor rank change due to perceived importance
3.7.4 Questionnaire Overall Risk Factor Ranking
According to the QU, the six highest ranked Risk Factors (weighted or un-weighted) were:
Speed;
Manoeuvring Space;
Tow Planning;
Training;
Tug Handling.
-4
-3
-2
-1
0
1
2
3
4
5
Ran
k C
han
ge
Risk Factor
48
3.8 Relationship between Risk Factor and Consequence
Severity
3.8.1 Pearson’s r Significant Number Test
A Pearson’s r significant number test, of the relationship between individual Risk Factor and
Consequence Significance, identified a Medium relationship (r value greater of 0.3) for three
factors:
Speed;
Tug Type;
Tow Planning
The test showed a Small relationship(r value greater than 0.1) for a further ten Risk Factors,
with no relationship found in six cases (See Table 23).
Risk Factor Pearson’s Number Correlation
Interaction 0.175452 SMALL
Girting 0.06754 NONE
Tow Planning 0.30077 MEDIUM
Tug Handling 0.28599 SMALL
Speed 0.37426 MEDIUM
Ship Size 0.1764 SMALL
Ship Power 0.21717 SMALL
Tug Type 0.32135 MEDIUM
Manoeuvring Space 0.16393 SMALL
Swell -0.0125 NONE
Current -0.1247 SMALL
Wind 0.06615 NONE
Visibility -0.0281 NONE
Ship Securing Arrangements 0.12329 SMALL
Equipment / Maintenance 0.03599 NONE
Communication Equipment 0.09308 NONE
Communications 0.15399 SMALL
Human Factors 0.15536 SMALL
Training 0.17561 SMALL
Total Number of Risk Factors 0.27939 SMALL
Table 23: Pearson’s r Significant Number Values
49
3.8.2 Risk Factor frequency compared to event significance
The largest positive rank movement (fourteen positions) resulting from the Pearson’s r test
was for Tug Type, while the largest negative movements were for Equipment / Maintenance
(nine positions) and Manoeuvring Space (eight positions).
Tow Planning, Speed, and Tug Handling remained top four Risk Factors; while Manoeuvring
Space, Human Factors and Training, fell in importance (See Graph 42).
Graph 41: Risk Factor Rank Movement: Questionnaire Frequency Versus Pearson’s r
Number
-2
1
-4
-9
0 0
5
0 0
-6 -8
2
-4
3 3 2
-1
14
4
-15
-10
-5
0
5
10
15
20
Ran
k M
ove
me
nt
Risk Factor
50
3.8.3 Relationship between Risk Factor Total Frequency and
Consequence Significance
A plot of consequence significance against Risk Factor total frequency indicated a small
correlation between the two qualities (See Graph 43.).
Graph 42: Risk Factor Frequency variation with Consequence Significance
A Pearson’s r test of correlation between Consequence Significance and Risk Factor total
frequency produced a value of 0.279, also indicating a Small relationship (value 0.1 to 0.3).
0
2
4
6
8
10
12
14
16
18
20
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31
Ris
k Fa
cto
r Fr
eq
ue
ncy
or
Co
nse
qu
en
ce
Sign
ific
ance
(n
um
be
r)
Questionnaires (Rank Ordered by Consequence)
Consequence Significance
Risk Factor Frequency
51
4.1 Suitability of Method
While the Case Study (CS) sample was limited to three states, the Questionnaire (QU)
provided a wider sample. In addition, although the Expert Interviews (EI) were limited to five,
convergence was observed, indicating a degree of saturation.
Use of a Judgement (rather than Random) sample of experts, with a total of one hundred
and ninety seven years relevant experience, helped extend understanding of issues;
additional written depositions and task observation, provided further validation.
Interpretative subjectivity concerns were managed by carefully following an objective
process; preconceptions were avoided by completing analysis of the interview data before
commencing other surveying.
As a qualitative analysis process, difficulties quantifying Grounded Theory data were
reduced by:
attaining saturation;
cataloguing code volume;
logging interviewee emphasis;
using follow-up to check specific points.
While the Questionnaire enabled collection of specific contemporary primary data, the issue
of authentication was managed by follow-up confirmation. Self-selection and normal
distribution problems, were reduced by use of non parametric tests.
The Qualitative and Quantitative processes had complementary aspects. The interviews
developed emergent theory and provided depth & granularity, but posed difficulties for
quantification; the surveys, using predetermined indicators enabled quantification, but risked
omitting symptoms (as demonstrated by the Questionnaire not including eight Risk Factors).
4.0 Discussion
52
Although interviews and surveys did not use identical codes, loss of comparative accuracy
was reduced by preserving generic categories so far as possible; use of three methods with
different data sources, also increased diversity. Triangulation, combining quantitative with
qualitative surveying, validated and deepened understanding of relationships between
variables.
This methodology helped identify and quantify Risk Factors in harbour towage operations,
enabling production of a baseline to help develop further analysis. A longitudinal
methodology might however improve evaluation of Risk Factor frequency against likelihood,
and help gauge success of any interventions.
53
4.2 Harbour Towage versus Non Harbour Towage
More frequent Collisions in harbour towage (HT) operations, compared with more frequent
Groundings in non harbour towage (NHT) operations, suggest the presence of differing
underlying Risk Factors; for example, Collisions might indicate Manoeuvring Space, while
Groundings might point towards Watchkeeping Risk Factors.
While harbour and non harbour towage had similar frequencies of towed Barges, harbour
towage also included a range of vessel categories. This difference may have been because
tugs involved in non harbour towage, had accidents where no other vessel was involved.
This non harbour towage characteristic was repeated, in high proportions of Unspecified
Bow Forms and Unknown Deadweights; and it compares with a broader cross section of
categories for harbour towage accidents. In this respect, harbour towage operations
accidents are more likely to involve another vessel.
With respect to Risk Factor variation, eleven factors were present in HT, but absent from
NHT operations; these Risk Factors may therefore be considered specific to HT operations.
A further five showed a more than twenty percent deviation; these Risk Factors may be more
common in HT operations.
Certain Risk Factors were present in over fifty percent of HT operations; their presence and
high frequency might be specific to these operations. By contrast other Risk Factors had
comparable or high frequencies in both groups, and might therefore be typical of all types of
operations. Others were comparatively more frequent in NHT, and may therefore not be
features of HT operations.
Only nine out of twenty eight Risk Factors had less than ten percent variation between NHT
and HT operations; they may therefore be present in similar frequencies in both types of
operation.
Comparatively higher total frequencies of Risk Factors in HT (See Graph 26) might indicate
volume of Risk Factors is a feature.
A Chi Square test of Risk Factor correlation between NHT and HT rejected the Null
Hypothesis in one hundred percent of cases; supporting the possibility that these two types
of operation can be differentiated.
54
4.3 Case Study versus Questionnaire Data
In both surveys, likelihood of Collision was twice the size of other categories (even with a
difference of twenty eight percent between them); by contrast there was a small difference in
likelihood of Damage.
There is a fifty five percent difference in the likelihood for Pollution between the surveys;
apart from this, categories retain similar proportions. A possible cause of this disparity is the
difference in type of Safety Event: Questionnaires comprised fifty five percent Near Misses
(where concern for pollution was perhaps higher than real risk, due to protective measures in
ship design).
Variation between survey data may also suggest two separate populations were studied;
over fifty percent of the QU concerned Medium powered ASD tugs, towing a range of vessel
types and Deadweights, involved in Near Misses. By contrast over fifty percent of CS
concerned Accidents to Moderately powered Conventional tugs, towing Barges of under ten
thousand tonnes deadweight. Weighting for Flag tug fleet size might however, allow a more
accurate evaluation.
While two different harbour towage populations may have been studied in the QU and CS,
similar Risk Factor profiles were observed. Although likelihood of Collision varied between
data sets, as a proportion of its cohort this difference was limited to eleven percent. Equally,
the largest difference was consequence of Pollution; however, as a proportion of their
respective cohorts, this amounted to a difference of seventeen percent.
The surveys identified thirty seven percent average frequency for Loss of Life, compared
with a fifty five percent average Injury frequency. There may be a number of reasons for this
including survey data sources only capturing substantial accidents.
Another potential source for the observed differences are disparities in the incident
categories. Ninety five percent of CS concerned Accidents, compared with nine percent of
QU: it is possible that this event seriousness may therefore have had an impact.
55
Differences in Vessel Type and Tow Position between data sets also require further
exploration. One possibility for observed variation is sample disparity; QU data was
collected from fourteen states, whereas two thirds of the CS were from North America,
where there is a well developed inland towage industry. Another potential explanation is
sample size; there were almost twice as many CS as QU.
Although, for eight Risk Factors, there was a difference of more than fifteen percent between
HT data sets, the majority more closely correlated; including the five most frequent Risk
Factors:
Tow planning;
Manoeuvring space;
Training;
Tug Handling;
Human factors.
Differences in frequency between the two HT surveys (& excluding NHT) led to an average
Risk Factor rank position movement of three places. The most profound Risk Factor
movement (Ship Power) was because it wasn’t included in the Case Study Data; however
the four most frequent Risk Factors remained the same in both surveys:
Human Factors;
Tow Planning;
Manoeuvring Space;
Training.
High frequency Risk Factors, identified in CS and EI, but not the QU, included:
Management Systems (eighty seven percent);
Following Operational Procedures (forty eight percent);
Passage Planning (forty four percent);
Watchkeeping (forty one percent);
The QU enabled a weighting to be added to individual Risk Factors to illustrate their effect in
an individual incident. While this amplified Risk Factor significance by different amounts
(and caused an average rank movement of one position) the six most significant Risk
Factors remained the same.
56
The largest rank change amplification and rank movement (four places) was for Speed. This
reflects the importance placed on Speed as a Risk Factor, by participants both in QU and EI.
This conclusion is supported by results from the Pearson’s r significant number test, which
produced the highest correlation (value 0.37426) between Risk Factor and Consequence
Significance, for Speed.
This test also supported the importance of Tow Planning and Tug Handling, and inflated the
significance of Tug Type; it however lowered the ranking of Manoeuvring Space, Human
Factors and Training. This may have been because these factors are more frequent and
therefore less important, to an individual accident.
A plot of Risk Factor volume against significance of incident, also indicated a relationship;
the higher the volume, the greater the accident significance. A Pearson’s r test value of
0.279 supported this (signifying a Small relationship between the two variables). This
feature should however be treated with caution, since individual Case Studies demonstrate
that one or two Risk Factors can be involved in some of the most catastrophic incidents;
while several Risk Factors can be implicated in lesser incidents:
Case study One: a tug pushing barges veered off course, ramming a pier and
collapsing a bridge, leading to multiple fatalities. The investigating safety agency
determined that the six risk factors were involved.
Case Study Two: a tug towing barges veered off course striking a dock and small
moored vessels. The investigating safety agency identified double the number of risk
factors, in an accident with fewer consequences.
57
4.4 Risk Factor Quantity
Since the data was not normally distributed, to test whether Safety Factor quantity had any
influence, a simple plot of accident severity against Safety Factor frequency showed some
increase, however there were:
significant fluctuations;
notable maximum Risk Factor frequencies in mid ranked incidents;
reductions in Risk Factor frequency for the most severe incidents.
A Pearson’s r test correlating Safety Factor total frequency with incident severity, produced a
value of 0.27939, signifying a Small relationship.
4.5 Risk Factor Severity
While analysis for presence of specific Risk Factors produced results in Graph 36, these
indicate volume of occurrence; they do not show the magnitude of their effect. For example,
while Safety Management Systems failure was observed in 87% of accidents, this Risk
Factor may have only played a small part as a cause; by contrast girting was only present in
17% of accidents investigated, but may have formed a more fundamental cause.
Therefore while Risk Factor frequency could provide an indication of its prevalence, this
does not indicate its significance in a particular event. The Pearson’s r test correlating Risk
Factor with perceived significance provided some indication of quality. For fifteen Risk
Factors this test and the frequency table, provided corresponding results; this may suggest
Risk Factor frequency also provides some indication of quality. There were two notable
exceptions, Manoeuvring Space was relegated to a ‘Small’ effect, while Tug Size was
increased to a ‘Medium’ effect.
58
4.6 Individual Risk Factor Explanation
Tow planning:
Was the next most frequent Risk Factor (seventy three percent). Although risks from
a lack of Tow Planning were not restricted to harbour towage, they were seven times
more frequent in these operations.
Manoeuvring Space:
One of five most frequent Risk Factors in the studies (sixty nine percent) it was not
specific to harbour towage, but eight times more frequent in these operations. The
Pearson’s r test correlation with consequence significance, gave it one of the largest
negative rankings.
Speed:
Although one of most frequently occurring risk factors, indentified in all three
surveys, and present in over half of events, it showed clear data variation. It was not
specific to harbour towage, but had a noticeably higher frequency during these
operations. Of all the Risk Factors, Pearson’s r test gave Speed the highest
correlation to consequence significance.
Human Factors:
Although one of the five most frequent Risk Factors in the studies, it was not specific
to harbour towage operations. While cited on seventy five percent of occasions,
there were mitigating statements indicating positive change. Human Factors may
have an elevated presence due to a particular need for tug crews to remain alert and
focussed. The Pearson’s r test showed a Small correlation with consequence
significance.
Safety Management Systems & Legislative Oversight:
Had the most significant frequency (eighty seven percent) but were not specific to
harbour towage operations. Expert Interviews coded this topic as one of the most
pressing in order to improve safety.
59
Training:
Also one of the most frequent Risk Factors in the studies, it was not specific to
harbour towage operations. The Pearson’s r test showed a correlation with
consequence significance. Training was cited on an average of sixty two percent of
occasions and had links with other Risk Factors such as Tug Handling, Crew Qualities
and Tow Planning and Communication.
Tug Handling:
One of the most frequent Risk Factors (fifty six percent) for harbour towage; the
Questionnaire weighting increased its importance by ninety four percent, while the
Pearson’s r test identified a correlation with consequence significance.
Following Operational Procedures:
Had a notable frequency (forty eight percent) but it was not specific to harbour
towage operations. Expert Interviews, reported that, “tugs have got such good
stability, but they won’t bounce back if you’ve got a door open”.
Rope Management & Seamanship:
Coded a signature Risk Factor for harbour towage operations (thirty four percent)
due to the frequency of working with ropes and the critical effects for safety of any
rope failure.
Interaction:
A Risk Factor for harbour towage operations (thirty nine percent) received a ninety
three percent weighting importance in Questionnaires; the Pearson’s r test signifying
a correlation with consequence significance.
Girting:
A signature Risk Factor for harbour towage operations (thirty three percent
frequency). The Pearson’s r test gave no indication of a correlation with
consequence significance; however Expert Interviews pointed out that it still
regularly occurs.
60
Tug Type:
Was a signature Risk Factor for harbour towage Operations (twenty two percent).
The Pearson’s r test identified Tug Type as one of three factors possessing a Medium
correlation with consequence significance.
Communications:
A notable Risk Factor in all three surveys (forty one percent) but not specific to
harbour towage; the Pearson’s r test showed a correlation with consequence
significance.
Tug Equipment / Maintenance:
Identified as a Risk Factor (thirty six percent) but not specific to harbour towage.
Linked with the significance of Safety Management Systems, Expert Interviews
emphasised that as a last line of defence, “tow hook Emergency Quick Release
mechanisms, not being maintained, have been the cause of many accidents”.
Communications Equipment:
Identified in thirteen percent of Harbour Towage operations, received a one hundred
percent ‘importance’ weighting increase in questionnaires.
Ship Securing Arrangements:
Specific to harbour towage operations (twenty eight percent); the Pearson’s r test
showed a Small correlation with consequence significance.
Ship Size:
A signature Risk Factor for harbour towage operations (thirty two percent); the
Pearson’s r test provided a Small correlation with consequence significance, however
there was clear variation within the data.
61
Ship Power:
Since it was not identified in the Case Studies this Risk Factor had a lower average
frequency (twenty three percent) giving a rank decrease of eleven places. It had a
Questionnaire ‘importance’ weighting increase of one hundred and thirteen percent
and the Pearson’s r test indicated a correlation with consequence significance.
Swell:
At twenty two percent this Risk Factor was not specific to harbour towage. The
Pearson’s r test gave no indication of a correlation with consequence significance,
although the Expert Interviews gave it increased weighting.
Current:
At twenty six percent, and as a signature Risk Factor to harbour towage operations,
the Pearson’s r test indicated a Small correlation with consequence significance.
Wind:
While not specific to harbour towage (twenty three percent frequency) Expert
Interviews increased its categorization.
Visibility:
Although at under ten percent frequency and not specific to harbour towage, Case
Studies did reveal a proportion of accidents occurred at night.
Crew Numbers:
Not specific to harbour towage operations (ten percent frequency) a number of crew
related risk factors concerning general purpose manning and appropriate
qualifications were identified by Expert Interviews.
62
Watchkeeping:
Had a frequency of forty one percent, but was not specific to harbour towage.
Passage Planning:
Had an average frequency of forty four percent, but was twenty two percent greater
in non harbour towage operations.
Bridge / Equipment Design:
It was not identified as specific to harbour towage (seventeen percent); Expert
Interviews however commented that new vessels can have dead slow speeds in
excess of ten knots, posing problems for securing tugs.
PPE:
Had a frequency of thirteen percent in harbour towage operations.
Other Risk Factors:
A range of other factors were reported including, working while intoxicated, and
poor tow marking.
63
5.1 Suitability of method
The methodology employed demonstrated a degree of success identifying and quantifying
Risk Factors in harbour towage operations. Triangulation, combining quantitative with
qualitative surveying, and contrasting emergent with established concepts, helped validate
and extend understanding of underlying relationships between the different variables.
The research enabled production of a baseline for further analysis; however greater use of
longitudinal sampling might improve identification of any relationship between Risk Factor
frequency and event likelihood, and help to measure success of any intervention.
5.0 Conclusions
64
5.2 Analysis of Risks to Safety
The results indicated that the Risk Factors involved in harbour towage operations could be
distinguished from those existing in other maritime sectors; however they also revealed that
the relationship between variables was complex, and that development of any solutions may
require considerable thought.
The Chi Square Test comparing harbour with non harbour towage operations rejected the
null hypothesis, indicating that the two sectors could be differentiated.
Analysis of the two sectors recognised the existence of different Risk Factors, and in
different proportions. Eleven Risk Factors were present only in harbour towage operations,
with a further three noticeably more frequent. Additionally, the average number of risk
factors was greater for harbour towage (nine) that for non harbour towage operations (five).
Particular Risk Factors specific to or with elevated frequencies during, harbour towage
operations, included:
Tow Planning (seventy five percent);
Manoeuvring Space (sixty three percent);
Tug Handling (fifty three percent);
Rope Management & Seamanship (thirty four percent);
Interaction (thirty two percent);
Speed (twenty seven percent).
Girting (twenty five percent);
Current (twenty four percent);
Risk Factors with significant frequencies (but not specific to harbour towage) included:
Management Systems (eighty two percent);
Human Factors (seventy eight percent);
Watchkeeping (fifty seven percent);
Passage Planning (fifty five percent);
Training (fifty four percent);
Following Operational Procedures (fifty percent).
Weighting of Risk Factor with significance of effect altered rank position, however the most
frequent Risk Factors remained the same.
65
A Pearson’s r test correlating Risk Factor with consequence significance, also altered rank
position, elevating relative importance of Tug Type, Ship Size & Power and Interaction.
A plot of Risk Factor frequency against consequence significance, indicated a potential
relationship, and a correlation test endorsed this.
Key Harbour Towage operations Risk Factors, identified by frequency or significance
1 Safety Management Systems (eighty two percent);
2 Tow Planning (seventy five percent);
3 Manoeuvring Space (sixty three percent);
4 Speed (twenty seven percent).
5 Human Factors (seventy eight percent);
6 Training (fifty four percent);
7 Tug Handling (fifty three percent);
8 Following Operational Procedures (fifty percent).
9 Rope Management & Seamanship (thirty four percent);
10 Interaction (thirty two percent);
11 Girting (twenty five percent);
12 Tug Type (twenty two percent);
13 Communications (forty one percent);
14 Tug Equipment & Maintenance (thirty six percent).
Table 24: Key Harbour Towage Risk Factors
Evidence of a correlation between Risk Factor frequency and consequence significance
should be balanced with the complexity of relationships between variables (and evidence
that a small number of Risk Factors can underlie the most catastrophic of accidents).
The most likely harbour towage operations safety event is Collision (seventy three percent)
followed by Capsize / Foundering (thirty two percent) and Grounding (twenty four percent).
The most likely consequence is Damage (ninety one percent) followed by Injury (sixty five
percent) and Pollution (thirty eight percent). There is also evidence of a prominent risk of
Loss of Life (thirty seven percent).
Variation between Case Study and Questionnaire data suggest two separate populations
may have been sampled (although no weighting for Flag State tug fleet size was used).
66
Over fifty percent of the Questionnaires concerned Medium powered ASD tugs, towing a
range of Vessel Types and Deadweights, involved in Near Misses. By contrast over fifty
percent of Case Studies concerned Accidents to Moderately powered Conventional tugs,
towing Barges of under ten thousand tonnes Deadweight.
Areas identified by the surveys that might benefit from further analysis include, provision of
an equitable system of regulatory oversight for the benefit of all tugs. A disproportionately
high number of smaller uninspected tugs, involved in accidents, provided limited anecdotal
evidence to support this. More substantial Risk Factor evidence includes the relative high
frequency of inadequate Safety Management Systems and Human Factors (legislated for in
International Maritime Conventions).
Changes to vessel design and increased complexity were identified as factors. Expert
Interviews reported that modern engine management systems can provide dead slow
speeds of ten knots; equally tug power has increased to an extent where bollard strength
can be insufficient.
The importance of an adequate number of appropriately qualified and experienced crew
were also recognised. A traditional method of training tug crews from deck boy to skipper, is
sometimes being replaced by migration from other maritime sectors; these new entrants may
not be similarly aware of Risk Factors specific to harbour towage operations.
Training issues underlay several Risk Factors: Tow Planning highlighted the importance of
the provision of sufficient information and experience to those managing harbour towage
operations; Following Operation Procedures underlined the importance of effective tug crew
drill programmes; and Tug Handling emphasised the importance of adequately preparing tug
masters.
Expert interviewees indicated that Training matters might also extended to personal qualities
and attitudes; pointing out the importance of teamwork and effective communication in safe
harbour towage operations. One particular code, Judgement (the ability to make a decision
concerning whether an action was safe) was highlighted; whether this was with respect to
handling a new generation of tugs (with reported exceptional tug size to power ratios) or the
ability to decide if a vessel’s speed was safe to close on her bow to make a tow.
67
6.1 Harbour towage operations reporting and research
This may help to monitor risks to safety and assist in developing solutions.
6.2 Equitable Regulatory Oversight and Monitoring
This may help provide an equitable system of regulatory oversight and monitoring, allowing
relevant International Conventions to benefit more tugs, while ensuring adequate Safety
Management Systems are in place.
A: Tug Masters and Crew: participate in confidential hazardous event
reporting.
B: Maritime Academic Institutions: undertake a comprehensive
detailed study of harbour towage operations safety risk.
A: Regulators: Review the current five hundred Gross Tonne threshold
in relevant International Conventions.
6.0 Recommendations
68
6.3 Enhanced Training Provision
Enhanced training provision may enable better planning and management of harbour tow
operations; common tug handling standards may help to build best practice; and, enhanced
harbour towage operation emergency drills may help to improve safety, for all those involved
in this crucial maritime sector.
A: Harbour Masters and Maritime Pilots Associations’: Consider
opportunities for improving harbour towage operations planning and
management.
B: Regulators and tug organisations: Consider development of common
standards of training in Tug Handling.
C: Regulators and tug organisations: Consider delivery of standard tug
harbour towage operations emergency drills.
69
ATSB, 2006. Investigation Report. Marine Occurrence Investigation. No. 224. Global Peace
and Tom Tough, 24 January 2006. [online] Available:
http://www.atsb.gov.au/media/24293/mair224_001.pdf [accessed: 10/02/13]
ATSB, 2011. tug Adonis, Capsize of the Australian registered tug Adonis at Gladstone, Qld,
11 June 2011. ATSB, 2013. [online] Available: http://www.atsb.gov.au/media/4077117/mo-
2011-005_final.doc [accessed: 12/02/13]
Australia Transport Safety Board (ATSB), 2011. 2005-2010 Australian Shipping occurrence
statistics (table 12). [online] Available:
http://www.atsb.gov.au/media/2495979/mr2011003.pdf [accessed: 10/03/13]
BTA (British Tugowners Association) 2010. Incident summary. [online] Available:
http:// www.britishtug.org%2FBTA_2010_5.pdf&ei=rbVyUKyON--
20QWu64DwDA&usg=AFQjCNFHqrrGGuEp-Zyvonbiy8_6Hdyk9A
[accessed: 10/10/12]
BTA (British Tug Association), 2010a & 2010b. Girting leads to constructive total loss; And,
Tug collides with another vessel. BTA Safety Seminar, Glasgow 4 November 2010. Incidents
and Safety Guidance Notice. Carthusian Court, London EC1M 6EZ
BTA (British Tug Association), 2012. 11th Annual Safety Seminar, 1st November2 2012.
Calman L, 2011. Manchester University. What is Grounded Theory? School of Nursing,
Midwifery and Social Work. University of Manchester. [online] Available:
http://www.methods.manchester.ac.uk/events/whatis/gt.pdf [accessed: 14/02/13]
Canada Shipping Act, 2001. Government of Canada. [online] Available:
http://www.tc.gc.ca/eng/acts-regulations/acts-2001c26.htm [accessed: 10/01/13]
CTSA (Canadian Transport Safety Agency), 2012. Shipping safety statistics. [online]
Available: http://www.tsb.gc.ca/eng/stats/marine/prelim-2011/index.asp [accessed: 17/02/13]
7.0 References
70
Charmaz K, 2006. Constructing Grounded Theory: A practical Guide Through Qualitative
Analysis. Sage. London.
Dand I W, 1975. Some Aspects of Tug Ship Interaction. Paper A5, in Proceedings of the
Fourth International Tug Convention, New Orleans, La London. Ship and Boat International.
Daniel W & Turner I, 2010. Qualitative Interview Design. The Qualitative Report, Vol. 15,
No.3.
Dick B, 2005. Grounded theory: a thumbnail sketch. [online] Available:
http://www.sce.carleton.ca/faculty/tanev/TTMG_5004/Articles/Dick_Grounded_theory_a_thu
mbnail_sketch_2005.pdf [accessed: 20/02/13]
DSB, 2010. Fairplay 22. [online] Available: http://www.fairplay.co.uk/ [accessed: 16/02/13]
EMSA, 2012. Marine incidents. [online] Available: http://emsa.europa.eu/marine-casualties-
a-incidents/occupational.html [accessed: 09/02/13]
ERM, 2012. Secondary data, statistical testing and correlation. [online] Available:
http://www.erm.ecs.soton.ac.uk/theme6/secondary_data_sources.html [accessed:
10/09/12]
ETA (European Tugowners Association) 2012. [online] Available:
http://www.eurotugowners.com/launch.cfm [accessed: 10/09/12]
ETA (European Tug Owners Association) 2011. Joint ETA-EMPA Guidelines On Design And
Layout Of Harbour Towage Equipment. [online] Available:
http://www.eurotugowners.com%2Fmedia%2Fmisc_media%2F110215%2520Best%2520Pr
act%2520Final%2520Rev%252011.pdf&ei=iheBUPvsCcOR0QW0k4G4BQ&usg=AFQjCNH
hCic6QT4LES96rGUkDxBVvnCp-w[accessed: 15/09/12]
European Harbour Masters Committee, 2010. Newsletter 19/03/2010
Excel Spreadsheets, 2013. Microsoft Office. [online] Available:
http://office.microsoft.com/en-gb/excel/ [accessed: 10/02/13]
71
Glaser BG & Strauss AL, 1967. The Discovery of Grounded Theory: Strategies for
Qualitative Research. Transaction. New Jersey.
Hallberg L R-M, 2006. The “core category” of grounded theory: Making constant
comparisons. 2006, Vol. 1, No. 3 , Pages 141-148 (doi:10.1080/17482620600858399).
School of Social and Health Sciences, Halmstad University, Sweden.
Henson H. 2003. Tug Use in Port. The Nautical Institute. London.
Henson H, Merkelbach D & van Wijnen FJ, 2011. Safe Tug Procedures. Seaways, June
2011. Nautical Institute.
Henson H, 2012. Safe tug operation: Who takes the lead? Tug & OSV, ABR company, July
2012.
Hensen H, Merkelbach D & van Wijnen FJ, 2013. Report on Safe Tug Procedures. 20 April
2013. [online] Available:
http://www.nvkk.nl/files/6113/6689/0145/Summary_All_Questionnaires_Final_version.pdf[ac
cessed: 16/02/13]
Hoda R, Noble J & Marshall S, 2011. Grounded Theory for Geeks. School of Engineering
and Computer Science. Victoria University of Wellington, New Zealand. [online] Available:
http://www.hillside.net/plop/2011/papers/E-13-Hoda.pdf [accessed: 11/11/12]
Holtzhausen S, 2001. Lancaster University. Triangulation as a powerful tool to strengthen
the qualitative research design: The Resource-based Learning Career. [online] Available:
http://www.leeds.ac.uk/educol/documents/00001759.htm[accessed: 10/09/12]
HSE, 2001. Root cause analysis. [online] Available:
https://www.hse.gov.uk/research/crr_pdf/2001/crr01325.pdf&sa=U&ei=oQSAUNmNAsnDhA
ekw4CYBw&ved=0CAcQFjAA&client=internal-uds-
cse&usg=AFQjCNFOHOvXpYvIDpK5HDAsHW31U6Bdjg[accessed: 18/09/12]
HSE, 2012 A. Risk Assessment. [online] Available: http://www.hse.gov.uk/risk/faq.htm
[accessed: 20/09/12]
72
ILO, 2013. International Labour Organisation: Maritime Labour Convention 2006. [online]
Available: http://www.ilo.org/global/standards/maritime-labour-convention/lang--en/index.htm
[accessed: 10/03/13]
IMO 1966. International Maritime Organisation: International Convention on Load Lines
Adoption: 5 April 1966. [online] Available:
http://www.imo.org/About/Conventions/ListOfConventions/Pages/International-Convention-
on-Load-Lines.aspx [accessed: 20/02/13]
IMO, 1974. SOLAS, 1974. International Convention for the Safety of Life at Sea (SOLAS),
1974. [online] Available:
http://www.imo.org/About/Conventions/ListOfConventions/Pages/International-Convention-
for-the-Safety-of-Life-at-Sea-(SOLAS),-1974.aspx [accessed: 15/02/13]
IMO, 1978. International Convention on Standards of Training, Certification and
Watchkeeping for Seafarers (STCW) Adoption: 7 July 1978. [online] Available:
http://www.imo.org/About/Conventions/ListOfConventions/Pages/International-Convention-
on-Standards-of-Training,-Certification-and-Watchkeeping-for-Seafarers-(STCW).aspx
[accessed: 04/02/13]
IMO, 2013. ISM regulations. [online] Available: http://www.imo.org/Pages/home.aspx
[accessed: 10/10/12]
International Maritime Pilots Association, 2013. [online] Available: http://www.impahq.org/
[accessed: 11/02/13]
International Tug and OSV magazine, 2012. [online] Available:
http://www.tugandosv.com/about_the_magazine.php [accessed: 18/02/13]
International Tugmasters Association, 2012. [online] Available:
http://www.tugmasters.org/books.html [accessed: 10/02/13]
Intute, 2006. Internet detective. [online] Available:
http://www.vtstutorials.ac.uk/detective/about.html [accessed: 25/09/12]
KMSB (Korean Maritime Safety Bureau) 2012. Shipping safety statistics. [online] Available:
http://www.kmst.go.kr/eng/cms/cms.asp?code=DA [accessed: 06/02/13]
73
Kunze B, 2011. BTA Seminar, The Human element. 10/11/2011.
Livingstone G and Livingstone G. 2006. Tug use offshore in Bays and Rivers: the
Towmaster’s manual. The Nautical Institute. London.
Livingstone G, 2012. The International Pilot. Issue 33, December 2012. Girting and Tripping.
Loughborough, 2010. Questionnaire structure. [online] Available:
http://coin.lboro.ac.uk/library/skills/quesdesign.html [accessed: 15/10/12]
MAIB, 2008. Marine Accident Investigation Branch. Flying Phantom tug incident report
(17/2008). [online] Available:
http://www.maib.gov.uk/publications/investigation_reports/2008/flying_phantom.cfm
[accessed: 18/09/12]
MAIB, 2012. Chiefton Tug incident report (12/20121). [online] Available:
http://www.maib.gov.uk/publications/investigation_reports/2012/chiefton.cfm
MAIB, 2013. Marine Accident Reports. http://www.maib.gov.uk/home/index.cfm [accessed:
06/03/13]
MAN, 2007. Propulsion trends in tankers, MAN, Copenhagen. 2007. [online] Available:
http://mandieselturbo.com/files/news/filesof11535/Propulsion%20trends%20in%20tankers.ht
m.pdf [accessed: 10/02/13]
Manchester University, 2011. Grounded Theory coded analysis. [online] Available:
http://www.methods.manchester.ac.uk/methods/groundedtheory/index.shtml [accessed:
18/09/12]
Marshall M N, 1996. Sampling for qualitative research. Family Practice. Oxford University
Press 1996. Vol. 13, No. 6. Printed in Great Britain. [online] Available:
http://spa.hust.edu.cn/2008/uploadfile/2009-9/20090916221539453.pdf [accessed: 16/02/13]
MCA, 2013. [online] Available: http://www.dft.gov.uk/mca/ [accessed: 18/09/12]
MSN 1767 (M) 2002. [online] Available: http://www.dft.gov.uk/mca/msn_1767_(m)-3.pdf
[accessed: 5/09/12]
74
Merchant Shipping (MS) (Load Line) Regulations 1998. United Kingdom Government.
[online] Available: http://www.legislation.gov.uk/uksi/1998/2241/contents/made [accessed:
10/02/13]
MS Regulation 1561 (1998). [online] Available: http://www.dft.gov.uk/mca/qms-cd-si-
1561.htm [accessed: 15/09/12]
MSN 1767 (M) Hours of Work, Safe Manning and Watchkeeping, Revised Provisions from 1
September 2002. [online] Available:
http://www.dft.gov.uk/mca/msn_1767_(m)-2.pdf [accessed: 20/02/13]
Merchant Shipping Act 1995 (c. 21). 1995 c. 21. [online] Available:
http://www.dft.gov.uk/mca/mcga07-home/shipsandcargoes/mcga-
shipsregsandguidance/mcga-mnotice.htm?textobjid=0A362EAE98415C02 [accessed:
15/02/13]
New Zealand Transport Accident Investigation Commission, 2000. Mahia. [online] Available:
http://www.taic.org.nz/ReportsandSafetyRecs/MarineReports/tabid/87/ctl/Detail/mid/484/Inv
Number/2000-208/Page/9/language/en-
US/Default.aspx?SkinSrc=[G]skins/taicMarine/skin_marine [accessed: 07/02/13]
New Zealand Transport Accident Investigation Commission, 2001. Tekomihana.
Investigation 01-204 Tug "Nautilus III", capsize and sinking, Auckland Harbour, 9 March
2001. [online] Available:
http://www.taic.org.nz/ReportsandSafetyRecs/MarineReports/tabid/87/ctl/Detail/mid/484/Inv
Number/2001-204/Page/9/language/en-
US/Default.aspx?SkinSrc=[G]skins/taicMarine/skin_marine [accessed: 10/02/13]
NTSB, 2002. National Transportation Safety Board. PB2004-916205. U.S. Towboat Robert
Y. Love. Interstate 40 Highway Bridge, Oklahoma. May 26, 2002. [online] Available:
http://www.ntsb.gov/doclib/reports/2004/HAR0405.pdf [accessed: 10/02/13]
Port Marine Safety Code, 2012. United Kingdom Government. [online] Available:
https://www.gov.uk/government/publications/port-marine-safety-code [accessed: 19/02/13]
75
The port of Heysham, 2011. Minimum Safety Standards for Tug Boats Operating within the
Jurisdiction of the Port of Heysham. [online] Available:
http://www.portofheysham.com/assets/pdf/Tug-Safety-Heysham.pdf [accessed: 20/02/13]
Sapsford R, 1999. Survey Research. Sage. London.
Statutory Instrument (SI) 2006 No. 3223 The Merchant Shipping (Inland Waterway and
Limited Coastal Operations) (Boatmasters' Qualifications and Hours of Work) Regulations
2006. [online] Available: http://www.dft.gov.uk/mca/mcga-
mnotice.htm?textobjid=658B5DDD96FCE317 [accessed: 10/02/13]
Slesinger J, 2010. ASD Tugs: Thrust and Azimuth. Cornell Maritime press. 978-0-87033-
617-1
Social Research Methods, 2006. Likert Scaling. [online] Available:
http://www.socialresearchmethods.net/kb/scallik.php [accessed: 25/09/12]
Solent University, 2012. Ethics Release Checklist for Research and Enterprise. [online]
Available:
http://learn.solent.ac.uk/file.php/75/Course_Documents/ethics_release_checklist_example.p
df [accessed: 21/09/12]
Stockman C, 2010. International Tugmasters Association. The importance of ‘being
prepared’ as demonstrated by a Melbourne tugboat almost girted - tug ‘Stockton II. [online]
Available: http://www.tugmasters.org/Melbourne%20tugboat%20almost%20girted.pdf
[accessed: 06/02/13]
Strauss A & Corbin J, 1994. Grounded Theory Methodology. [online] Available:
http://cms.educ.ttu.edu/uploadedFiles/personnel-folder/lee-duemer/epsy-
5382/documents/Grounded%20theory%20methodology.pdf [accessed: 10/10/12]
Tellis W, 1997. Application of a Case Study Methodology. The Qualitative Report, Volume 3,
Number 3, September, 1997. [online] Available: http://www.nova.edu/ssss/QR/QR3-
3/tellis2.html?ref=dizinler.com [accessed: 24/11/12]
Tokyo MoU on PSC in the Asia Pacific region. [online] Available:
http://www.tokyo-mou.org/ [accessed: 10/02/13]
76
Transport Safety Board of Canada (TSBC), 2013. 2011 Marine Annual Statistics. [online]
Available: http://www.tsb.gc.ca/eng/stats/marine/2010/ss10.asp#sec1 [accessed: 19/02/13]
TSBC (Transport Safety Board of Canada), 1995. M95W0020. Marine Occurrence Report.
Tug "Seacap XII". Fraser River, 11 May 1995. [online] Available:
http://www.tsb.gc.ca/eng/rapports-reports/marine/1995/m95w0020/m95w0020.pdf
[accessed: 25/10/12]
UK Harbour Masters Association, 2012. [online] Available: http://www.ukhma.org/
[accessed: 10/02/13]
United States of America (USA), US Title 33 (Navigation and Navigable Waters). [online]
Available: http://uscode.house.gov/download/pls/Title_33.txt [accessed: 15/02/13]
United States of America, US Title 46, U.S. Code. [online] Available:
http://www.navcen.uscg.gov%2Fpdf%2FAIS%2FAIS_Regs_SOLAS_MTSA_FR.pdf&ei=VBI
ZUfXpA7ON0wXwqoGoDQ&usg=AFQjCNEqV8Bt0HQMS8bdQmxMiL5WLQLcDg&bvm=bv.
42080656,d.d2k [accessed: 12/02/13]
University of Southampton, 2012. e Research Methods, 2012. Sources of secondary data.
[online] Available: http://www.erm.ecs.soton.ac.uk/theme6/secondary_data_sources.html
[accessed: 10/02/13]
USCG, 2009. Homeland Security. Marine Safety Information Bulletin #03-09. April 3rd 2009.
Reducing Downstreaming Incidents on the Western Rivers.
USCG, 2012. News: Salvage operations of Arthur J and Tug Madison begin. [online]
Available: http://www.uscgnews.com/go/doc/4007/1500915/Salvage-operations-of-Arthur-J-
and-Tug-Madison-begin-most-Lake-Huron-beaches-reopen [accessed: 18/02/13]
A
Appendix A.
A-1 Example Tug Towage Safety Questionnaire
This questionnaire has been produced by Stephen Ford, a serving tugmaster, as
part of a research project into Tug Safety, with Solent University. The purpose of
this questionnaire is to identify the specific risks involved in harbour towage
operations.
It is addressed to all tug masters and asks them to describe one tug towage job
which raised a safety issue. This might have been a near miss or an incident, but it
includes challenging jobs; for example where you did things differently the next
time.
All information provided will remain strictly confidential to myself, the researcher.
Anything that attributes information to a particular person, vessel or company will
be removed: any information provided is purely for statistical analysis. All
responses will be destroyed upon completion of the research project in June 2013.
To fill in the form, tug masters are asked to put a cross (X) in the most applicable
box. If a question is not applicable, you are unsure about the answer, or you do
not wish to answer a question, please leave it blank.
Please email your completed questionnaire by 10/04/2013 to:
For more information about the project and for the results of the research
(available June 2013) please go to:
http://mahara.solent.ac.uk/view/view.php?id=66091
Alternatively I can be contacted at the above email address.
Thank you for your help with this research.
B
Details of the tug towage job which raised a safety issue
The Tug 1. Type of tug Conventional
(propeller/ rudder)
ASD Tractor Not known / other
………
2. Tug Bollard Pull approximate (tonnes)
Moderate <30t
Medium 31t - 65t
High > 66t
Not known / other
The Assisted Vessel 3. Assisted vessel type (Container, barge, etc.)
(Please state) … …………………………………………………..
4. Approximate size (deadweight, tonnes)
Barge or Coaster (<10,000t)
Handy (10,000-50,000t)
Large (51,000-160,000)
Very Large (>160,000)
Not known / other
5. Bow form Fine Moderate Broad Unsure / other (Please state)
……
Tug Assistance Provided 6. Tug Help Push &/or pull Tow on a
line Not known / other
(Please state)
………
7. Tug Position
Tug Forward Amidships Tug Aft Not known / other (Please state)
………
8. Whose line
Tug’s Ship’s Unsure / Not Appropriate
External factors
Conditions / Weather 9. Wind (Beaufort Scale)
Low <F3
Moderate F4-F6
Gale F7-F83
Storm >F9
Unsure
10. Swell height (m)
Calm <0.2
Moderate 0.3 – 0.9
Rough 1.0-1.5
Storm >1.5
Unsure
11. Current (knots)
Low <1
Moderate 2-3
Strong >3
Unsure
12. Other external condition (please state)
……………………………
C
The Safety Issue
The risk 13. How likely were the following outcomes in this case
Not likely / Not applicable
Possible Likely Highly likely Inevitable
a. Collision
b. Grounding
c. Foundering
d. Major damage
e. Minor damage
f. Loss of life
g. Major Injury
h. Minor injury
i. Pollution
The event 14. How would you best describe the event:
Challenging / Instructive
Near miss Incident Accident Other
Brief description 15. Please use this space to briefly describe any other factors not already covered elsewhere.
D
Thank you for completing this questionnaire; please email it to: [email protected]
Extent to which different factors were involved
Factors involved 16. Indicate the extent to which the following factors influenced the safety issue.
Factor No effect / Not Applicable
Some effect Important effect
Fundamental effect
Interaction between tug & tow (e.g. tug enters bow pressure wave)
Girting, Girding or Tripping (e.g. gobbing down gear absent or inadequate)
Insufficient detailed planning of tow (e.g. insufficient tug bollard pull ordered)
Ship speed too fast (e.g. tug unable to maintain required position)
Ship size too large for tug/s (e.g. windage too great for tug bollard pull)
Ship too powerful for tug/s (e.g. main engines overpowering tug pull)
Wrong tug type used for job (e.g. conventional tug used in vulnerable position)
Lack of manoeuvring space (e.g. restricted by shallow water, buoys or piers)
Excessive swell (i.e. snatching or parting tow line)
Excessive current (e.g. difficult to control tow)
Excessive wind effect (e.g. too great for tug power)
Practical difficulties (e.g. need to remain in ‘critical area’ to pass tow line)
Poor ship securing arrangements (e.g. poor fairleads or bollards)
Tug equipment (e.g. failure of emergency quick release)
Inadequate communication equipment (e.g. poor VHF)
Language difficulties (e.g. lack of spoken English)
Human factors (e.g. poor concentration, fatigue, etc.)
Training insufficient (e.g. not enough time on new tug type)
Other (please state) ………………………..