Implementing an Effective Safety Culture

1. Introduction

Safety at sea has improved enormously over the last 100 years, and continues to do so.

Rules and regulations, beginning with SOLAS have given the industry a compulsory framework to follow. Furthermore with the increase in engineering and electronic technology, unexpected mechanical or structural failure is rare, and the influence of the elements becomes less and less of a factor.

However, in one five year period, 2869 commercial crew and passengers have lost their lives at sea1 . Safety at sea has developed through many phases, from the reactive seeking of guilt and apportion of blame, to compulsory adherence to rules and regulations.

It is now time to fully embrace a third phase: that of a “just-culture” and self-regulation. A positive safety culture depends on two factors, the development and implementation of a proportionate and suitable safety management system that reports-on and examines its own failings, and a positive culture at all levels so that management and crew truly understand that safety is in their best interests.

The bedrock of these two factors is training: Training arms senior management with the skill and knowledge required to develop a practical safety management system, and gives the crew the perspective and attitude to want to follow it.

2. Scope & Purpose

The purpose of this document is to discuss and investigate the meaning of “an effective safety culture”, the positive financial, moral and legal implications of implementing one, the potential threats, barriers and pitfalls precluding the successful adoption of an effective safety culture and, most importantly, offer guidance on how to overcome these obstacles. It is intended to be a frank and open discussion, to seek to educate and inform the general reader (without overcomplicating the issue), as well as stimulate dialogue with the health and safety professional.

It is offered as a free and open-source document in order to generate interest and raise awareness of a crucial and significant issue, one of great consequence to the maritime industry due to the perilous nature of living and working at sea, in a heavily industrialised context. It satisfies our ambition to fulfil our moral duty of care to the seafarer, who provides an increasingly vital service to human society, under increasing economic pressure, with ever more demanding expectations of efficiency, productivity and compliance. It forms part of a suite of educational, supportive and consultative services provided to the maritime industry.

3. Statistics

In order to effectively mitigate the effects of incidents, accidents and dangerous occurrences we must first investigate the nature and cause of these occurrences.

Above we can see a typical example for the types of occurrences from a Red Ensign Group Flag administration’s annual safety report (over 1000 registered vessels). At first glance it becomes obvious that the figures appear counter-intuitive. The figures are wildly over-representative in favour of serious, damaging or life threatening occurrences.

To examine the most obvious example: how is it possible to have 33% of the incidents involving “Collision, foundering, heavy contact or stranding” but only 1% of incidents involving a (“consequence free”) COLREG Infringement (See “Expanded” wedges in fig.1). The short answer is, “It isn’t”. The anticipated ratio of “consequential” incidents to near misses to dangerous occurrences, based on long term and exhaustive study across industry, is expected to be in the region of 15:1345 This tendency to under-report “near misses” is well documented, and is examined in detail in this document (See below, 7.1 “Reporting”).

The negative impact of this tendency, the underlying reasons for it, and solutions to mitigate this trend is one of the central premises of this document.

The consequences of under-reporting are made clear in the above chart.

While the frequency of accidents outnumbers the occurrence of casualties (as would be expected according to empirical evidence789 ), the number of near misses reported is proportionately lower than would be expected. Though this sample is too small to be statistically significant10, it is difficult not to notice (in an anecdotal sense) that the decline in the number of near misses reported is mirrored by a commensurate increase in the number of casualties.

With this problem in mind, and to further explore the causal relationship between Dangerous occurrences (or near misses) and damaging or deadly incidents and accidents, we must now investigate the pathology of an accident – namely “how and why do they happen”

4. Accident Causation

Accidents do not just “happen”11, and to think logically about how they happen is a useful preventative tool, if done pre-emptively as part of a risk assessment, or to prevent reoccurrence if done as part of an accident investigation.

Many Health and Safety studies cite the “Swiss Cheese” model12, with the layers of cheese representing preventative and control measures.

Only when the “holes” line up, do accidents occur. To expand on this, it is useful to think about the relative complexity of a task and significance of the consequences of failure. A relatively simple task, involving few people and with minor potential risk will be covered by a few (hopefully common sense) risk controls resulting from a simple risk assessment and/or pre task planning (see 6.3 “Planning and Implementation”).

A more complex task, involving a larger or more technically complex system, or an ongoing routine task implemented across a fleet or industry will be governed by increasingly more layers of safeguards:

This, “Fortuitous break in the chain of events13” would lead to a “hazardous occurrence” or “near miss” rather than causing harm. (See 7.1 “Reporting”). In a safety management system where only “harm” is recorded and measured, and near misses/hazardous occurrences are overlooked or under-reported for any reason, (see 7.1.1 “Barriers to Reporting”) the hazard goes unnoticed, and the failings (the fallen dominoes) remain unidentified and unrectified until it does, eventually, result in harm.

Full report: Best Management Practices for Implementing an Effective Safety Culture