Learning from the past to enhance the resilience of nuclear power plants against natural disasters

Abstract

Natural disasters pose a significant threat to critical infrastructures, particularly nuclear power plants (NPPs), where the impact of a natural hazard can have far-reaching consequences. That is why enhancing the resilience of NPPs, meaning the ability to prepare for, withstand, adapt to and recover from a disaster, is vital. This thesis aims to enhance resilience by strengthening the learning process from past disasters. And lay the groundwork for a learning process guideline document to be adopted by NPPs as an addition to their existing practices to strengthen their resilience learning. The existing resilience practices within NPPs tend to reactive rather than pro-active. With limited integration of structured learning mechanisms and digital technologies.

This research develops the Enhanced Learning Process (ELP) model. Designed to visualize a structured and digital technology integrated resilience learning approach. Employing a Socio-Technical Systems theory perspective to ensure alignment between technology adoption and organizational dynamics. The systematic literature review (SLR) performed examined the literature of resilience engineering, organizational learning and digital technology utilization. And identified existing gaps in resilience strategies, which included the underutilization of digital tools for historical data processing, simulated scenario-based learning and collaborative decision-making. The synthesized literature, analyzed from a socio-technical systems theory perspective, created the knowledge foundation to structure a theoretical enhanced learning process model proposed in this study. Leveraging digital technology capabilities from a socio-technical perspective to enhance historical disaster data processing, identifying possible underlying vulnerabilities of the plant and cascading effects in the event of a natural disaster. Serving as the foundation of a resilience learning guideline that nuclear power plants can adopt to strengthen their learning from the past capabilities to enhance their resilience practices. The proposed model consists of four stages (Data Collection, Data Analysis, Outcome Evaluation and Strategy Development). Reflecting the resilience learning process within NPPs. Where each stage leverages digital technologies to enhance learning capabilities and support the development of data-driven and sophisticated resilience strategies.

The validity and practical applicability of the model was evaluated through a case-based analysis of the Fukushima Daiichi nuclear disaster. Applying the proposed model to this case, demonstrated its’ applicability in a real-world scenario. And provided insights into points of improvement for their learning processes. Lastly, an expert in the industry was consulted to provide verification of the outcomes of this research and validation of practical application of the proposed model.

The outcomes of this study contribute to the fields of resilience engineering and organizational learning by proposing a transformation of traditional learning processes through a structured, digital technology supported, learning approach to enhance resilience strategies against natural hazards. Furthermore, this study extends the application of STS theory in resilience enhancement learning by structuring the learning process as a socio-technical system. Providing insights for researchers, policy makers and NPP operators seeking to strengthen the resilience and learning capabilities of NPPs. This study suggests further research to focus on further validating and developing the ELP-model in collaboration industry experts and further explore the influence that digital technologies have on learning capabilities.