Learning from incidents: Practices at a Scandinavian refinery
Introduction
Few would deny that learning is important in organizations that work with high-risk technologies in dynamic and volatile environments. Much research is devoted to understanding how this process can be expressed and measured, how the “right lessons”1 can be learned and how companies can facilitate more efficient and effective learning (see e.g. Drupsteen et al., 2013, Hovden et al., 2011, Jacobsson et al., 2011). Organizational and safety research has identified a variety of factors that either inhibit or facilitate learning from incidents (see e.g. Pidgeon and O’Leary, 2000, Schilling and Kluge, 2009, Smith and Elliott, 2007, Størseth and Tinmannsvik, 2012).
But there is still much to learn about the actual practice of incident learning. We do not really know how learning “happens” in the day-to-day work in high-risk industries. What we need are empirical studies of the practices by which organizational members seek to learn from incidents and try to incorporate lessons in their working routines (Lindberg et al., 2010).
One industrial sector in which these processes may be studied is the oil and gas industry. Learning in this sector is both critically important and very hard. Recent disasters (think of the Deepwater Horizon disaster in 2010) underline the importance of learning (Skogdalen, 2011). At the same time, the BP Texas City, Tesoro Anacortes, Longford and Buncefield accidents have shown that it may not be easy to learn the right lessons.2 The petroleum processing and refining industry is a highly competitive globalized industry with large, and, in Europe, often aging production facilities (Wood et al., 2013). The industry has to balance the need to reduce costs with the necessity to improve safety performance.
In this paper, we present findings from an exploratory study of a Scandinavian refinery. We wanted to study a refinery where there was a systemic emphasis on learning from incidents. We chose our particular refinery in a country where the regulation regime and national supervisory body prioritize learning from incidents and accidents at petroleum-related facilities.
We selected this particular refinery because it was the largest refinery in the host country and because it had been the site of several serious incidents during the 2000s.3 In the context of the European refinery industry, it is a medium-sized refinery. The refinery consists of a natural gas liquids fractionation plant, a crude oil terminal and crude storage facilities, a combined heat and power plant and a refinery with a refining capacity of 12 million tons per year. The refinery used a safety barrier approach and risk-based tools as part of its safety management (Vinnem, 2007).
This organization was a real “learning lab” given the many incidents it had to handle. The company that owned and operated the refinery granted access to the site and all relevant documentation. The refinery organization had been criticized by the supervisory authority for not doing enough to learn from incidents following an audit in 2010. However, the selected refinery was working to improve learning from incidents, and was willing to allow the lead author of this paper access to its operations and employees.
The research questions were straightforward: What do practices aimed at learning from incidents consist of, how are they carried out and what are their results? In this article we describe how employees understood and experienced activities to learn from incidents as part of their working lives. Our research is based on analyses of extensive fieldwork and interviews with 70 employees at the refinery. We start by discussing theoretical perspectives on incident learning. We then present the findings of our fieldwork. After discussing our findings, we conclude with an overview of implications for practice and theory.
Section snippets
Theoretical perspectives on learning from incidents
Organizations that work with complex, high-risk technologies in volatile environments cannot afford to wait for a crisis or disaster to happen and then start learning. They must exploit every opportunity to learn. Every incident or near-miss represents such an opportunity.
Organizations that do this all the time and continue to perform successfully are described as being High Reliability Organizations (HROs). In recent years, we have seen the rise of these HROs in the safety literature (see e.g.
Studying learning practices
Studies of the literature on incident learning describe a need for empirical studies (Le Coze, 2013, Lindberg et al., 2010, Drupsteen and Guldenmund, 2014). Our study is based on extensive fieldwork, making use of observations and interviews. It provides insight into the variety of processes that make up efforts to learn in a single organization. It shows – despite an organization’s attempt to control and direct learning efforts – how practices, and the way employees experience them, can vary.
Learning practices at the refinery
We found that practices aimed at learning from incidents took place both in a structured, formal way (using an incident management system), as well as in other, less structured and more informal ways. Below we first describe who participated in activities aimed at learning. We then describe the nature of these activities. Finally, we consider how respondents perceived the results of their participation.
Discussion
Research on incident learning in the safety and organizational sciences tends to examine instrumental processes that do or do not lead to learning (see e.g. Crossan et al., 1999, Kjellén, 2000, Kolb, 1984, Ramanujam and Goodman, 2011, Størseth and Tinmannsvik, 2012). These process theories of learning make universal claims regarding the suitability of certain types of learning process and lesson types.
We understand learning from incidents as processes of knowledge development, which may
Conclusion: Lessons for practice and research
We were fortunate to gain access to a large refinery, which allowed us to study practices of learning in the wake of incidents. We used a practice-oriented approach, focusing both on formal and informal learning processes. Our findings yield some intriguing insights that may be beneficial to others who study incident learning, in academia or practice.
Learning from incidents is not exclusively governed by a safety or incident management system. It can also be realized by individuals and groups
Role of the funding source
Statoil, the owner of the refinery where the research for this article was carried out, funds the PhD project that this article is a part of. The company has had no involvement in determining the nature or extent of data collection, interpretation and analysis of collected data, in determining the content of the article or the decision to submit the article for publication.
Acknowledgements
We wish to thank Statoil for funding the research that is presented in this article and for allowing access to their refining facility. We are grateful to all employees who allowed us to shadow them during their workday and who were interviewed as part of this project.
References (54)
- et al.
Components of a tool to address learning from accident investigation
Int. J. Disaster Risk Reduct.
(2013) - et al.
Whatever happened to process theories of learning?
Learn., Cult. Soc. Inter.
(2012) - et al.
Multilevel learning from accidents – case studies in transport
Saf. Sci.
(2011) - et al.
Method for evaluating learning from incidents using the idea of “level of learning”
J. Loss Prev. Process Ind.
(2011) What have we learned about learning from accidents? Post-disaster reflections
Saf. Sci.
(2013)- et al.
Learning from accidents – what more do we need to know?
Saf. Sci.
(2010) - et al.
Man-made disasters: why technology and organizations (sometimes) fail
Saf. Sci.
(2000) - et al.
The challenge of collective learning from event analysis
Saf. Sci.
(2011) - et al.
Situated learning and education
Educ. Res.
(1996) - Ansell, C., Boin, A. (Eds.), 2011. Special Issue. In Honor of Todd R. LaPorte. J. Contingen. Crisis Manage., vol. 19,...
Organizational Learning: A Theory of Action Perspective
Organizational Learning II: Theory, Method, and Practice
Knowledge and the theory of organizations: organizations as activity systems and the reframing of management∗
J. Manage. Stud.
The practice of risk governance: lessons from the field
J. Risk Res.
The legacy of the high reliability organization project
J. Contingen. Crisis Manage.
Organizational Learning: The Interplay of Routinization and Crisis in Practices
Constructing Grounded Theory: A Practical Guide through Qualitative Analysis
Through the practice lens: where is the bandwagon of practice-based studies heading?
Manage. Learn.
An organizational learning framework: from intuition to institution
Acad. Manage. Rev.
What is learning? A review of the safety literature to define learning from incidents, accidents and disasters
J. Contingen. Crisis Manage.
Critical steps in learning from incidents: using learning potential in the process from reporting an incident to accident prevention
Int. J. Occupat. Safety Ergon.
Can governments Learn?: American Foreign Policy and Central American Revolutions
From Organizational Learning to Practice-Based Knowing
Human Relat.
On claims that answer the wrong questions
Educ. Res.
Cited by (24)
Learning from cyber security incidents: A systematic review and future research agenda
2023, Computers and SecuritySystemic constraints to effective learning from incidents in the Ghanaian mining industry: A correspondence analysis and AcciMap approach
2020, Safety ScienceCitation Excerpt :However, what has been written so far focuses on specific stages of the learning from incidents process, while other stages seem to have been neglected. For example, several factors have been identified that confront various aspects of incidents reporting including, non-detection/non-identification of reportable incidents (Drupsteen and Hasle, 2014; Lindberg et al., 2010; Sanne, 2012) and underreporting of detect incidents (Drupsteen et al., 2013; Lukic et al., 2012; Lukic et al., 2010; Russell Vastveit et al., 2015; Sanne, 2008; van der Schaaf and Kanse, 2004). Similarly, others have identified factors affecting the search for causes (Harms-Ringdahl, 2009; Hollnagel, 2008; Lindberg et al., 2010; Lundberg et al., 2009; Ramanujam and Goodman, 2011).
Public leaders’ organizational learning orientations in the wake of a crisis and the role of public service motivation
2019, Safety ScienceCitation Excerpt :Here, the four dimensions that we identified require further rigorous testing in new contexts – in terms of agents, organizations, and institutional settings – to build confidence in the measurement instrument. This study further contributes to the literature by addressing and specifying the political dimension of crisis-induced organizational learning in addition to a common ‘technical’ approach to the process (e.g., Choularton, 2001; Vastveit et al., 2015; Silva et al., 2017). Crisis-induced learning inherently differs in several respects from organizational learning in more ‘regular’ times.
Investigating work-related musculoskeletal disorders: Strengths and weaknesses of current practices in large Australian organisations
2019, Safety ScienceCitation Excerpt :This suggests that in many organisations, improving the investigation process is only one of various changes required to improve WMSD prevention. This supports findings from other domains, which have shown that formal investigation processes are only a small part of the processes required for learning (Lundberg et al., 2010; Ramanujam and Goodman, 2011; Vastveit et al., 2015). Whilst key issues were identified, some practices were also identified that are likely to significantly enhance opportunities for learning.
Failure to learn from safety incidents: Status, challenges and opportunities
2018, Safety ScienceCitation Excerpt :Lundberg et al. (2012) identified situations where investigators were scared of critiquing management although that was the right thing to do. Concerning the same issue, participants in an interview conducted by Russell Vastveit et al. (2015) has this to say “no, it is rare to write something about leadership…maybe we do not have a culture for it, so you could say that the incident system largely has been aimed at the operators” (p. 84). In the same study, another participant remarked “it is rare to write [measures] aimed at leadership or the HSSE department; mainly they [the measures that the respondent produced] are technical measures that can be aimed at the maintenance department” (p. 84).
Learning from incidents in health care: Critique from a Safety-II perspective
2017, Safety ScienceCitation Excerpt :Their empirical model was validated and refined against semi-structured interview data from members of two multinational corporations in the energy sector (a refinery and a gas plant). Despite these advances described in the literature, empirical work suggests that substantive difficulties still persist across industries, and that organisations are struggling to transform knowledge about vulnerabilities into change in practice (Drupsteen and Hasle, 2014; Anderson and Kodate, 2015; Le Coze, 2013; Sujan, 2015; Pilbeam et al., 2016; Vastveit et al., 2015). The studies suggest that weaknesses exist in all steps of the learning cycle affecting both the breadth and the depth of learning (Lukic et al., 2012).