A sequential method to identify underlying causes from industrial accidents reported to the MARS database

The application of systematic accident analysis tools to investigate food safety incidents

Effective food safety (FS) management relies on the understanding of the factors that contribute to FS incidents (FSIs) and the means for their mitigation and control. This review aims to explore the application of systematic accident analysis tools to both design FS management systems (FSMSs) as well as to investigate FSI to identify contributive and causative factors associated with FSI and the means for their elimination or control. The study has compared and contrasted the diverse characteristics of linear, epidemiological, and systematic accident analysis tools and hazard analysis critical control point (HACCP) and the types and depth of qualitative and quantitative analysis they promote. Systematic accident analysis tools, such as the Accident Map Model, the Functional Resonance Accident Model, or the Systems Theoretical Accident Model and Processes, are flexible systematic approaches to analyzing FSI within a socio-technical food system which is complex and continually evolving. They can be applied at organizational, supply chain, or wider food system levels. As with the application of HACCP principles, the process is time-consuming and requires skilled users to achieve the level of systematic analysis required to ensure effective validation and verification of FSMS and revalidation and reverification following an FSI. Effective revalidation and reverification are essential to prevent recurrent FSI and to inform new practices and processes for emergent FS concerns and the means for their control.

Advancing Hazard Assessment of Energy Accidents in the Natural Gas Sector with Rough Set Theory and Decision Rules

The impacts of energy accidents are of primary interest for risk and resilience analysts, decision makers, and the general public. They can cause human health and environmental impacts, economic and societal losses, which justifies the interest in developing models to mitigate these adverse outcomes. We present a classification model for sorting energy accidents in the natural gas sector into hazard classes, according to their potential fatalities. The model is built on decision rules, which are knowledge blocks in the form of “if (condition), then (classification to hazard class x)”. They were extracted by the rough sets method using natural gas accident data from 1970–2016 of the Energy-related Severe Accident Database (ENSAD) of the Paul Scherrer Institut (PSI), the most authoritative information source for accidents in the energy sector. This was the first attempt to explore the relationships between the descriptors of energy accidents and the consequence (fatalities). The model was applied to a set of hypothetical accidents to show how the decision-making process could be supported when there is an interest in knowing which class (i.e., low, medium, high) of fatalities an energy accident could cause. The successful use of this approach in the natural gas sector proves that it can be also adapted for other energy chains, such as oil and coal.

Failing to Fix What is Found: Risk Accommodation in the Oil and Gas Industry

The present program of research synthesizes the findings from three studies in line with two goals. First, the present research explores how the oil and gas industry is performing at risk mitigation in terms of finding and fixing errors when they occur. Second, the present research explores what factors in the work environment relate to a risk-accommodating environment. Study 1 presents a descriptive evaluation of high-consequence incidents at 34 oil and gas companies over a 12-month period (N = 873), especially in terms of those companies' effectiveness at investigating and fixing errors. The analysis found that most investigations were fair in terms of quality (mean = 75.50%), with a smaller proportion that were weak (mean = 11.40%) or strong (mean = 13.24%). Furthermore, most companies took at least one corrective action for high-consequence incidents, but few of these corrective actions were confirmed as having been completed (mean = 13.77%). In fact, most corrective actions were secondary interim administrative controls (e.g., having a safety meeting) rather than fair or strong controls (e.g., training, engineering elimination). Study 2a found that several environmental factors explain the 56.41% variance in safety, including management's disengagement from safety concerns, finding and fixing errors, safety management system effectiveness, training, employee safety, procedures, and a production-over-safety culture. Qualitative results from Study 2b suggest that a compliance-based culture of adhering to liability concerns, out-group blame, and a production-over-safety orientation may all impede safety effectiveness.

Why do organizations not learn from incidents? Bottlenecks, causes and conditions for a failure to effectively learn

If organizations would be able to learn more effectively from incidents that occurred in the past, future incidents and consequential injury or damage can be prevented. To improve learning from incidents, this study aimed to identify limiting factors, i.e. the causes of the failure to effectively learn. In seven organizations focus groups were held to discuss factors that according to employees contributed to the failure to learn. By use of a model of the learning from incidents process, the steps, where difficulties for learning arose, became visible, and the causes for these difficulties could be studied. Difficulties were identified in multiple steps of the learning process, but most difficulties became visible when planning actions, which is the phase that bridges the gap from incident investigation to actions for improvement. The main causes for learning difficulties, which were identified by the participants in this study, were tightly related to the learning process, but some indirect causes - or conditions - such as lack of ownership and limitations in expertise were also mentioned. The results illustrate that there are two types of causes for the failure to effectively learn: direct causes and indirect causes, here called conditions. By actively and systematically studying learning, more conditions might be identified and indicators for a successful learning process may be determined. Studying the learning process does, however, require a shift from learning from incidents to learning to learn.

Systems thinking, the Swiss Cheese Model and accident analysis: a comparative systemic analysis of the Grayrigg train derailment using the ATSB, AcciMap and STAMP models

The Swiss Cheese Model (SCM) is the most popular accident causation model and is widely used throughout various industries. A debate exists in the research literature over whether the SCM remains a viable tool for accident analysis. Critics of the model suggest that it provides a sequential, oversimplified view of accidents. Conversely, proponents suggest that it embodies the concepts of systems theory, as per the contemporary systemic analysis techniques. The aim of this paper was to consider whether the SCM can provide a systems thinking approach and remain a viable option for accident analysis. To achieve this, the train derailment at Grayrigg was analysed with an SCM-based model (the ATSB accident investigation model) and two systemic accident analysis methods (AcciMap and STAMP). The analysis outputs and usage of the techniques were compared. The findings of the study showed that each model applied the systems thinking approach. However, the ATSB model and AcciMap graphically presented their findings in a more succinct manner, whereas STAMP more clearly embodied the concepts of systems theory. The study suggests that, whilst the selection of an analysis method is subject to trade-offs that practitioners and researchers must make, the SCM remains a viable model for accident analysis.

Systemic accident analysis: examining the gap between research and practice

The systems approach is arguably the dominant concept within accident analysis research. Viewing accidents as a result of uncontrolled system interactions, it forms the theoretical basis of various systemic accident analysis (SAA) models and methods. Despite the proposed benefits of SAA, such as an improved description of accident causation, evidence within the scientific literature suggests that these techniques are not being used in practice and that a research-practice gap exists. The aim of this study was to explore the issues stemming from research and practice which could hinder the awareness, adoption and usage of SAA. To achieve this, semi-structured interviews were conducted with 42 safety experts from ten countries and a variety of industries, including rail, aviation and maritime. This study suggests that the research-practice gap should be closed and efforts to bridge the gap should focus on ensuring that systemic methods meet the needs of practitioners and improving the communication of SAA research.