Data Science and Business Intelligence Techniques for Learning from Environmental Accident Analysis for Offshore Oil Fields

Abstract

Incorporating data science and business intelligence (BI) techniques as a strategy and tool for improving and evolving process safety for the oil and gas industry is a no-return method that should provide extraordinary gains. The technology is a powerful and necessary partner for the oil industry to overcome the challenges of new frontiers for oil exploration and production. Additionally, this applies to the health, safety, and environmental segment of business because more challenging scenarios imply greater potential risks; therefore, access to information within the appropriate time, clearly, and consistently allows the longevity of business. Digital transformation, helped current activities supported by weak instruments (i.e., spreadsheets, e-mails, etc.) to migrate to a database structure that facilitated the understanding of their real situation within the appropriate time—at macro and micro levels—allowing adequate support for decision-making. BI tools aided by data science techniques facilitate decision-making, often extracting productive information from content-rich texts. The combination of data science techniques with BI tools enables a full-blown experience for business analysts through new insights, background connections not yet discussed, more professional visualizations, and telling the same or a new story using more complete, and often complex, innovative questions and answers. Answering essential questions for process safety in the oil and gas industry when analyzing environmental accidents, atmospheric dispersions, emissions, leaks, and spills in a structured method (presenting graphically within the context of rigs), multiple views of the problem allow improved management of efforts, which reduces the number of cases. The same concept can be expanded to questions related to injuries, machinery and/or equipment damage, performance, etc.