Upstream Oil and Gas
Enterprise-wide Analytics
A very large oil and gas company recently embarked on an ambitious, multi-phased digital transformation strategy to increase revenue by reducing production deferrals and running operations at more than 95 percent efficiency. Key to this was taking the billions of data points generated each year across its complex offshore oil and gas assets and making it available in a contextualized data lake.
With Element, the company is experiencing significant project time and cost savings. Project costs savings amount to $15.9m by building OSIsoft PI System AFs internally and without the cost of external/SI consultants (3 full time consultants for 6 months). The project was also delivered significantly earlier due to > 90% reduced effort in building AFs (from 6 months to 3 weeks).
Offshore assets face increasing pressure to reduce deferrals, maximize margins and reduce health and environmental incidents. Risk in upstream organizations is shifting from exploration to production, increasing expectations for running world class operations. Oil and gas leaders are adjusting to this new reality.
Unfortunately, the company’s digital transformation campaign fell behind schedule. The problem wasn’t data analysis but collecting and managing the data in an efficient, scalable, accurate way. In most cases, the company was already gathering and analyzing needed data, but the efforts were generally localized, inefficient and didn’t scale. Technicians and engineers produced and maintained endless spreadsheets. Or they developed bespoke applications for narrow, local objectives that didn’t help technicians elsewhere. For example, if a technician identified equipment in Africa that needed more frequent maintenance to run optimally, the technician in the Caribbean, overseeing the same equipment, couldn’t benefit from the insight.
Senior technician – “Getting data from other teams is hard. We’ll ask for compressor data, but it won’t come back for weeks, and is incomplete. That back and forth saps people’s energy and morale.” Without a 360-degree view of operations-related data, insights remain out of reach.
These challenges impacted three key use cases in phase 1 of the transformation:
Element’s FDEs and customer SMEs worked closely with business leaders to ensure that data collection and management were aimed at producing the biggest business impacts.
The following activities were performed:
Element Unify hosted within customers Azure tenant
Ingest metadata from OSIsoft PI data archives, SAP Maximo, hazard data, P&ID diagrams, and engineering data sheets
Design and build equipment-centric Asset Twins based upon the desired attributes of their target equipment
Transform and contextualize process, maintenance and safety data
Export metadata in the form of PI Asset Frameworks (AFs) to import back into the OSIsoft PI System
Export various hierarchies and raw data to the customers Azure storage for use with GE Predix APM, application developers, and data scientist
Element’s value is delivering trusted data from source to consumer so that we can reduce the amount of unplanned outages and deferrals, and maximize margin. They are helping us harmonize our data, and ensure its reliability, so we can use it with all of our applications in a trustworthy, repeatable fashion.
Element Unify is demonstrating the power of asset data modeling and management at scale. Key to the success is eliminating the manual and time-consuming work typically performed in spreadsheets to prepare data for analysis and industrial applications.
With Element Unify, the company is experiencing significant project time and cost savings. Project costs savings amount to $15.9m by building OSIsoft PI System AFs internally and without the cost of external/SI consultants (3 full time consultants for 6 months). The project was also delivered significantly earlier due to > 90% reduced effort in building AFs (from 6 months to 3 weeks).
Additionally, key company use cases benefited in the following ways:
Trying to get data from other teams is hard. We’ll ask for compressor signal data, but it won’t come back for weeks, and even then, it’s only half of what we need. That back and forth just saps people’s energy and morale.