LOSS OF CONTAINMENT INSIGHTS PROJECT - CLOUD ANALYTICS PLATFORM

The Discovering Safety Programme (DSP) is a large technical programme of work. The programme aims to extract safety insights from chaotic and unstructured free text data using Text Mining (TM) and Natural Language Processing (NLP) techniques. The Loss of Containment (LoC) Insights project is focused upon preventing high-consequence events in the onshore process industries. The project team have been using analytical software and storage on a legacy network providing limited software development and analytics capabilities. A key focus of this work has been using the RStudio package running in a Windows environment.

The data uploaded to the new cloud solution will consist largely of freetext incident reports and other sources of intelligence (e.g. inspection/audit reports). This data will be subject to automated anonymisation to remove personal and organisational data prior to upload. However, as a precaution the security arrangements will need to assume that the dataset still contains personal data.