Elsevier

Safety Science

Volume 45, Issue 6, July 2007, Pages 697-722
Safety Science

Implementing a safety culture in a major multi-national

https://doi.org/10.1016/j.ssci.2007.04.005Get rights and content

Abstract

This paper reports on the implementation of an advanced safety culture in a major oil and gas multi-national. The original proposal came from the company after it had become clear that expectations had been raised after the successful implementation of Health, Safety and Environment (HSE) Management Systems subsequent to the Piper Alpha disaster. The proposal made by the company, to develop a workforce intrinsically motivated for HSE, was operationalised as the development of an advanced safety culture after a review of the literature on motivation. The model used was the HSE Culture Ladder that had become the industry standard accepted by the OGP (International Association of Oil and Gas Producers). This model was intended to show that there were considerable opportunities for improvement even after HSE-MS had been implemented and that the more advanced cultures were ones people felt were desirable and achievable for themselves. Once top management had provided the initial support for the development of a more advanced safety culture, a number of supporting tools were developed, under the Hearts and Minds brand, and a strategy for implementation was developed that relied more on bottom-up ‘pull’ rather than top-down ‘push’ – the standard implementation model for new initiatives. The tools were designed to provide a clear direction, a road map to an advanced culture defined in terms provided by people within the industry, to support lasting changes in attitudes and beliefs, to promote an increased feeling of control when solving HSE-specific problems – all components of a more advanced culture. The tactics employed, using a pull rather than a push approach, had to allow for local variation within the general limits set by the strategy that eventually became a mixed top-down and bottom-up approach. Next there is a discussion of the current status and the lessons to be learnt from the implementation so far: moving away from command and control is hard for large organizations; such programs have to be driven by different performance indicators; managers have to learn to disperse their control; it is essential to communicate both successes and failures. Finally there is a discussion about the respective roles of academia and the industry in such endeavours, the requirement to concentrate on more than a single cultural characteristic such as reporting, and the difficulties of evaluating such programs in a worldwide environment that is continuously changing.

Introduction

The problem of safety culture – What is it and how do you become one? – is probably the main issue in modern thinking about safety (Turner and Pidgeon, 1997). Since the IAEA report (IAEA, 1991) on the Chernobyl disaster, which introduced the concept of a safety culture to a wider world, failures arising from the culture of an organisation have become seen as the reason why major accidents happen, such as the loss of the space shuttle Columbia (NASA, 2003) and many recent railway disasters such as Clapham Junction (Hidden, 1989), Ladbroke Grove (Cullen, 2001) and the Waterfall disaster (McInerney, 2005). Most attention has been paid to the issue of safety climate (Zohar, 1980), a concept easier to measure (e.g. Flin et al., 2000), but the underlying expectation is that the best and safest organisations have a culture of safety, and that safety climate is an indirect measure of how close an organisation approximates to that.

This paper reports on one program intended to implement a safety culture in a major multi-national oil and gas company. This is a different problem from implementing a safety culture in a single, physically distinct, organisation for a number of reasons. While implementing a safety culture in a small, bounded organisation can be done by the creation of a clear and simple vision and single-minded commitment to that vision from the top, large organisations are much harder to steer, even with full commitment. In a small organisation there will be a more restricted range of operations, leading to a restricted organisational structure; most people can know, or know of, each other. Even a large international company with a limited focus, such as a petrochemical company like DuPont or Dow Chemical, or an aircraft manufacturer like Boeing or Airbus, does not have all these issues to contend with. A modern vertically integrated Oil and Gas Major, like Shell, BP or Exxon, covers a wide range of activities including exploration, oil and gas production, shipping, refining, chemical production, transport and sales. Each of these distinct operations has their own hazards and, often, a set of histories that may result from long practice or be the results of the acquisitions that made them as large as they are. In multi-national organisations there will be a wide range of operations, people will be physically dispersed over a range of time-zones, will operate in different national settings and will also speak a variety of languages. In large organisations there will be a large number of sub-organisations, each with their own history, having a potentially distinct culture and run by managers with their own vision of where to go, and how. The sheer size of the problem and the very nature of such organisations mean that a close range hands-on approach is not feasible. Another factor that has to be considered is the interaction of national with organisational culture. The academic literature has concentrated primarily upon a limited number of types of organisation and studies have been typically carried out in Western environments (Guldenmund, 2000). We have little to guide us when we step outside the comfort zone, the Western cultural environment that has been studied in some detail.

This paper sets out the problem as initially posed by the company, as this was not originally an explicit request to change the culture, although it was soon reframed as such. Then follows an outline of the strategy for implementation that has been developed and that is being carried out, followed by a description of the tactics being employed to develop the culture in specific locations. I will then attempt to assess the difficulties encountered and evaluate the current status of what is still an early stage in the program – a 10-year lifespan is not an unreasonable estimate of how long it may take to acquire a highly developed culture.

Section snippets

Understanding the causes of accidents

The program to be described here forms the continuation of an earlier research program, for the same company, on understanding accident causation that went back as far as 1985. Initially intended to replace an accident taxonomy (Wagenaar and Hudson, 1986), this program developed into a model of how accidents happen (Wagenaar, 1986, Wagenaar and Groeneweg, 1987, Reason et al., 1988, Wagenaar et al., 1994, Reason, 1990, Reason, 1997) that has become widely known as the Swiss cheese model. Within

The culture model

The concept of a safety culture is frequently treated as part of a dichotomy, an organisation either is or is not such a culture; discussion has centred round issues such as whether an organisation has or is such a culture, and what are the characteristics that make such a culture (Hale, 2000, Guldenmund, 2000, Reason, 1997, Reason, 1998). If one merely wishes to describe such a culture this is a natural way to proceed, but should one wish actually to make an organisation become such a culture,

The strategy

The story here is written with the benefit of a degree of hindsight, but hopefully the lessons can serve to develop more foresight in future. Much of the strategy was developed as the program progressed, from frequent small field trials in many different parts of the world, followed by extensive examination of progress and what was identified as working, both well and badly. Most of the actual program was driven by the development of new tools and the accumulation of experience as a result of

The tactics

With the strategy laid out above the next stage involves learning how far back one can stand and how, and how often, one has to intervene physically to ensure that the program is not derailed or loses impetus. The strategy involving pull rather than push means that much control is taken out of the hands of the Centre and of the small group who developed the whole program. Strategy involved creating a product, an advanced HSE Culture, that people actively wanted and then providing them with the

Current status

There has been a considerable take-up and the Hearts and Minds program, in one form or another, is in operation in almost all parts of Shell Group. In 2004 implementation of the Hearts and Minds program became one of the HSE priorities for the Group. This priority was set by the Group HSE Committee that is chaired by Shell Group’s CEO. Almost all management teams have at least gone through the HSE Culture exercises, in which they identify where their operation is on the ladder and select one or

Conclusion

The question that needs to be faced is: Is it working? Is the culture being changed and is the approach described here effective? The answers to these questions seem to be generally positive, but in some ways it is too early to tell. Unlike the descriptions of the high reliability organisations this is not a state that has already been achieved, but rather one that is aspired to. Certainly every experience with asking people what kind of culture they would like has led to people selecting a

References (54)

  • The Hon. D. Cullen

    The Ladbroke Grove Rail Inquiry

    (2001)
  • A. Furnham

    Management and myths: challenging the fads, fallacies and fashions

    (2004)
  • A. Hidden

    Investigation into the Clapham Junction Railway Accident

    (1992)
  • A. Hopkins

    Safety, Culture and Risk: The Organisational Causes of Disasters

    (2005)
  • Hudson, P.T.W., 1991. Prevention of accidents involving hazardous substances: the role of the human factor in plant...
  • Hudson, P.T.W., Parker, D., 1999. Micro-tools for the management of change. Micro-tool Manual 1. Report for Shell...
  • Hudson, P.T.W., Parker, D., 2001. Profiling safety culture: the OGP study. Report for the Oil and Gas Producers...
  • P.T.W. Hudson et al.

    Cost and benefit in HSE: a model for calculation of cost–benefit using incident potential

  • P.T.W. Hudson et al.

    Tripod Delta: proactive approach to enhanced safety

    Journal of Petroleum Technology

    (1994)
  • Hudson, P.T.W., Verschuur, W.L.G., Lawton, R., Parker, D., Reason, J.T., 1997. Bending the rules II. Report for Shell...
  • Hudson, P.T.W., Verschuur, W.L.G., Parker, D., Lawton, R., 1998. Intrinsic motivation for HSE: The Hearts and Minds...
  • Hudson, P.T.W., Verschuur, W.L.G., Croes, S., Parker, D., Lawrie, M., 1999. Hearts and Minds. Report on Field Trial...
  • P.T.W. Hudson et al.

    The Hearts and Minds project: creating intrinsic motivation for HSE

  • P.T.W. Hudson et al.

    HSE tools: Which tools are appropriate?

  • Hudson, P.T.W., Verschuur, W.L.G., Parker, D., Lawton, R., van der Graaf, G.C., Kalff, J., 2000c. Bending the rules:...
  • P.T.W. Hudson et al.

    Managing non-compliance: moving from theory to practice

  • P.T.W. Hudson et al.

    The Hearts and Minds Program: understanding HSE culture

  • Cited by (213)

    • Learning not blaming: Investigating ten fatal road traffic collisions using STAMP-CAST

      2023, Transportation Research Interdisciplinary Perspectives
    View all citing articles on Scopus
    View full text