Achieving greater resilience to major events: Organisational learning for safety risk management in complex environments

Major events in industries where safety is dependent on sometimes complex interactions between engineered systems, processes, and human behavioural issues, still occur relatively frequently. This sometimes leads to loss of life and injury, environmental damage and financial and reputational penalties. This Report considers how organisational and cultural precursors to such events, along with associated behavioural issues, can be better identified and managed, and offers the potential to develop greater resilience to them.

It is divided into five Chapters. Following a general introduction, Chapter 2 presents the results from analysing the organisational and cultural learning identified from formal reports into twelve major events drawn from the petrochemical, nuclear, transport and civil engineering sectors. This analysis has identified a set of precursors (with high commonality between events) which have then been classified under ten 'themes' (leadership, culture, business pressures, learning, competence etc.). In order to facilitate application of these findings, the identified precursors were distilled into a set of 'Expectations' associated with each theme and supported by a discussion and commentary which puts them into context and draws out associated issues. It is proposed that organisations may wish to 'benchmark' their current equivalent material against these findings from real events, and identify any areas for improvement. Expectations and the associated discussion within each theme have deliberately been made self-standing, and although this leads to some overlap between themes, it will allow organisations to benchmark either collectively against all themes or selectively where particular concerns or vulnerabilities may have been identified. Further potential uses of the expectations can be found in Chapter 5.
The capacity to recognise threats and vulnerabilities is key to maintaining and enhancing organisational and constituent system resilience. Chapter 3 focuses on opportunities for organisational learning to produce a strategic perspective on intervention to engineer behavioural and cultural change. It centres on evidence gathering in the related domains of: identifying structural, organisational and socio-technical weakness; setting priorities for intervention to enhance resilience; intervention options appraisal – theories of change and ‘what works’? and, intervention design / delivery i.e. the intervention logic – how will ‘it’ work?

Recognising what needs to change, in behavioural and cultural terms can be challenging for work organisations, particularly where there is a desire to determine root causes. However, whilst difficult, this constitutes a less demanding objective compared to challenge of successfully socially engineering change in the workplace.

Following a discussion of the relative contribution of lag and lead data (and its treatment) to organisational learning, the Chapter moves-on to discuss the relative merits of headline contemporary behaviour change models and techniques. A core theme of the Chapter is its emphasis on the need to achieve a good fit between: (i) the change model / technique selected and the intervention objectives i.e. considerations relating to how impact will occur (theory of change), and (ii) the extent to which contextual variables embody the potential to facilitate or thwart the intervention (intervention logic).

The core themes of: measuring health and safety performance; techniques for priority elicitation; options for intervention; setting an agenda for improvement and considerations when designing behaviour change interventions are discussed with reference to established workplace safety and management science insights, and through reference to broader advances and consolidations of perspectives on behaviour change and its delivery, notably within the public policy domain. The body text is supported by a set of themed appendices on: high reliability organisations; safety culture maturity; strengths and weaknesses of influential behaviour change perspectives; training and communication interventions, and incentives and reward schemes.

Chapter 4 deals with the issue of how interventions can be designed which take account of the complex and interactive nature of many proposed actions. Put simply, an improvement in one area can lead to unexpected and unwanted consequences in others, unless systems aspects are adequately addressed. Specifically, it introduces Causal Loop Diagrams (CLDs) as a participatory tool which can be used to identify issues and to design and explore interventions in a way which seeks to explicitly understand the interacting issues which may occur in the design of potential interventions to promote improvement. It shows through a series of case studies how the technique can be used to 'prototype' proposed changes and explore their potential for introducing unwanted 'knock on' effects.

The recommendations in Chapter 5 suggest two ways in which the value of the expectations could be maximised. Firstly, it should be possible to develop from the expectations, sets of 'penetrating questions' that would allow organisations to determine the extent to which operational activity reflects the defined features of good practice. Secondly, since many event investigations still do not address in significant depth some of the organisational and cultural precursors to events, a new framework could be developed which would prompt deeper investigations where relevant. Similarly, there is reflection on ways in which CLDs can might be further developed and applied to combine consideration of the behavioural issues identified in Chapter 3 within the framework of CLD, potentially allowing, for the first time, an integrated approach to be taken of engineered and behavioural factors in the design of more effective interventions and more resilient systems.