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

Report Executive Summary

Phase 1 of the current study published by the EI in March 2021 involved an analysis of the organisational and cultural precursors to twelve major events across a range of industries and concluded that these were strikingly similar. The findings were brought together under ten ‘themes’ - leadership, business pressures, safety culture, communications, the safety management system, risk assessment and management, the ability to develop and maintain a learning organisation, competence, the management of contractors and oversight and scrutiny.
There are a variety of established techniques for assessing process safety risks arising from engineering causes, but less attention is often given to these organisational and cultural precursors, and there do not exist widely used techniques for assessing vulnerability and developing interventions which might be effective in reducing their potential impact. The approach suggested below should be applicable across a wide range of industries.
Developing an effective approach is seen to involve three steps:
The first is to provide organisational ‘expectations’ or ‘good practices’ which, if followed, would minimise the risks of organisational accidents. In the Part 1 Report, a comprehensive set of such expectations with an associated ‘commentary’ was developed based on the findings from the events studied and categorised under the ten themes listed above. It was suggested that organisations may wish to ‘benchmark’ their equivalent material against these ‘expectations’ or use them as a basis for developing such statements of expected good practice. They may also wish to augment the ‘expectations’ based on their own experience and/or organisational requirements, although the source of the current set means that they are believed to cover most of the organisational and cultural shortcomings which have been precursors to major events. It is intended that they should be updated if any new learning becomes available.
The second step involves the development of a systematic process to enable organisations to assess where the ‘expectations’ are not being met in practice. In this Phase 2 report, sets of questions and associated ‘statements’ have been developed which are designed to enable key groups in the organisation – leaders, ‘specialists’, and the operational workforce - to provide an assessment of the extent to which ‘reality’ aligns with organisational ‘expectations’. A suggested process for analysing output to identify potential vulnerabilities, and to prioritise actions to address these, has then been developed and is presented in Part 1 of this report.
Thirdly, it is vital to ensure that interventions address the identified shortfalls, and within a complex system, do not introduce undesirable and unintended ‘knock-on’ effects. A system modelling technique using causal loop diagrams (CLDs) was outlined in the Phase I report as a means to minimise such effects. In this Report, examples are provided of how these may be used to address findings from the application of the question sets, taking account of the potentially complex interactions which are likely to arise in practice. This approach should then enable the development of suitable measures (performance indicators) to measure the effectiveness of implementation of proposed interventions.
The Phase I Report also highlighted the importance of not only addressing shortcomings by making changes to organisational arrangements and improving the underlying safety culture, but also the great importance of considering potential attitudinal and behavioural impacts which might arise from proposed improvements. Examples were given of how these may develop. In the current report, these are given particular prominence in an example set of CLDs developed from findings identified in some of the events studied. This involves a failure to maintain stability in the teams which have the responsibility to manage and ‘drive through’ required changes to a satisfactory conclusion.
This analysis draws together understanding obtained from the systems engineering and social sciences expertise of the authors – a ‘linkage’ in thinking which has not always been available in addressing complex interactive organisational and ‘people’ based matters such as those that have been identified from the events studied. It is considered very important that the full range of potential precursors to serious accidents is addressed in all industries which rely on excellence in engineering, organisational processes, and in the performance of individuals and teams in sometimes complex and strongly interactive environments.
The intention is that the approach outlined might be subject to trial in organisations which see the importance of the matters covered by the two reports. In this way, it is hoped that a mature practical technique might be developed which can be widely applied.