As digitalization continues to transform the Oil and Gas industry, remote monitoring and surveillance is fast becoming a strategic focus area within many organizations. Data virtualization, visualization, and workflow orchestration are critical functions within this area. Major challenges for digitalization are how to make the best use of the combination of structured and unstructured data, which are often disconnected, as well as how to orchestrate the workflow for minimum friction and maximum efficiency.
This paper describes the successful development and implementation of a Digital Decision Assistant (DDA), a digital environment that integrates data ingestion from various sources, multiple micro-workflows, and bots that use machine learning and natural language processing to execute tasks and make captured knowledge available within and across teams. The DDA uses real time data ingestion, process automation, machine learning, and natural language processing all integrated into a singular adaptive platform which empowers end users by making relevant data available and enabling predictive analytics. Correlating time series data with condition-based monitoring alerts, and user instructions, comments, and tags enriches the data and provides necessary context. Search and workflow functions further extend the platform as both a knowledge repository as well as an orchestrator to outside applications and analytical tools. The platform also includes automated functionalities for; event detection, notifications, alarm management, exception-based surveillance, data preparation, linkage to other applications (via API's), key performance indicator (KPI) generation, and reporting.
The DDA suite provides an extensible and scalable remote monitoring and control system that can incorporate contextual information. The use of data virtualization, automation, and decision enabling tools within an integrated platform allow the organization to improve data virtualization, visualization, and workflow orchestration across a wide spectrum of real time applications, as well as enrich datasets for future digitalization initiatives.