Speaking at the PTN Events Oil and Gas Digital Transformation virtual/online conference, Rob Kennedy (Wood) advocated the application of ‘right to left thinking’ to ‘transform lifecycle economics’. Right to left means starting with what is needed at the end of the facility lifecycle and working back to design, allocating capex appropriately. For Wood, the process revolves around a full asset lifecycle methodology that includes building a digital twin ‘from the ground up’. Wood’s flagship digital twin was developed for Turkish Petroleum’s Sakarya Gas Field to accelerate project delivery, maximize operating performance and reduce costs. The ‘twin’ comprises an asset twin, GIS and process twin, built with component software from Hexagon, ESRI and Wood’s own Virtuoso asset performance management tool that is said to embed data standards based on CFIHOS and PODS.
Daniel Fleck (RedEye) described engineering information management (EIM) as an often-overlooked area of asset management. EIM involves multiple documents and data types, from 2D drawings, CAD files to 3D models and BIM* data, through images, documents and controlled records. Working with this disparate information remains hard. A survey revealed that engineering information mis-management is ‘significant and costly’. Redeye proposes a generic template for the development of a digital engineering standard that is comprehensive and future-proof for digital twin and BIM usage. EIM needs to be agnostic, supporting all platforms and tools natively. Enter Redeye’s Common Data Environment, a hosted, multi-tenant solution for workflow and engineering information management.
* Building information model/management.
Maria Coelho presented digital engineering work performed at DICE, the Digital Innovation Center of Excellence (an unit of INL, the Idaho National Laboratory) spanning model-based design, digital threads/twins, artificial intelligence and extended reality. INL, a national center for nuclear research is managed by the Battelle Energy Alliance. Model-based systems engineering is described as the application of modeling and data-driven engineering in support of systems-of-systems design, analysis and life cycle operations. MBSE transforms typical systems artifact documents to data objects, creating a ‘single source of truth’. MBSE informs the digital twin, a ‘living virtual model of the physical asset’ that is used to predict future behavior. Real-time bi-directional communication between the twin and the asset compares simulated and measured information. It is this integration of real-time data and dynamic model update that distinguished the digital twin from the ‘traditional’ simulator. Moreover, the digital twin blends sensors and instrumentation, artificial intelligence, and online monitoring into a ‘single cohesive unit*’. INL’s digital twins are built around DeepLynx (and on Git) a central data warehouse of live ‘ontological’ and time series data. AI is used to automate expensive and manual human activities and to predict unobserved and difficult to measure events.
* One might think that this is easier said than done!
Sarunas Strasevicius (StackFlows) claims that ‘60% of skilled worker time is wasted on coordination’ activities, responding to emails, chats, follow-up meetings and tracking down missing input. ‘Employees feel that two-thirds of scheduled meetings are unnecessary*’. ‘Over 10 percent of an employee’s day is spent on tasks that have already been completed, either by themselves or by a colleague.’ Strasevicius enumerated some misconceptions of digital transformation. In-house IT will never have the resources to realize the transformation. On the other hand, contrary to the legacy IT approach, modern tools can make a transformation ‘remarkably affordable’. The key here is across-the-board business process automation, led by the operations team who ‘know how processes should run’, and who are the best-equipped to achieve the transformation. Strasevicius recommends the use of the Object Management Group’s BPMN process modelling standard. The ISO standard modeling language is ‘easy and intuitive’ and can be learned in a month, along with the companion OMG Decision Model and Notation spec which adds rules and automation to BPMN. StackFlows integrational platform leverages all of the above with bi-directional integration of third party (ERP, CRM … ) systems.
* A statement that recalls John Wannamaker’s observation that ‘Half the money I spend on advertising is wasted; the trouble is I don't know which half.’
Matthew Moore is the global subject matter expert for condition monitoring at contractor Petrofac. He observed that predictive maintenance and condition monitoring in oil and gas is ‘very mature’ due to the exceptional return on investment associated with avoiding downtime and reducing maintenance costs. Petrofac has reaped the benefits of digitalization through its CBMnet reporting tool and advisory for critical rotating machinery. Petrofac is now working to progress condition monitoring with automated data collection and analysis, moving towards real-time data and the use of wearable technology (headset) and ‘facilitate the use of AI and machine learning’. Real time vibration data collection is getting attention today with the advent of IIoT wireless vibration sensors and cloud-based AI and ML. Condition monitoring has been ‘glamorized as an IIoT digital initiative and introduced to more industrial sectors, particularly manufacturing. The market is now ‘hugely competitive’ due to the availability of low cost MEMS technology. Moore walked through some rather intricate failure cases to show that while the AI/ML approach has a lot to offer, the new technology brings new challenges and increased costs. ‘Condition Monitoring is set to become significantly more expensive, but will value scale proportionally?’ ‘AI and ML for prescriptive maintenance is still some way off but remains the ultimate prize’.
Kai Eberspaecher works for Bengal Energy, a Canadian operator exploring in the Australian outback. He presented a ‘day in the life’ of a field operator, an activity that is digitally-enabled with a range of off-the-shelf applications. The operator’s 200 mile drive is tracked with a journey management system from Brisbane-based JMS. Driver fatigue and heat exposure is monitored with Canaria. Work orders are managed with Redeye (see above) and captured in Rockwell Automation’s Factory Talk Fiix computerized maintenance management (CMMS) package. Data from the Ignition Scada/historian system is analyzed in Energysys’ production accounting. Bengal’s digital transformation design principles are ‘off-the-shelf, cloud-based and value for money’.
Johnpaul Portelli presented Canada Natural’s clean resource innovation network (CRIN). In 2020, the Alberta government instituted FEMP, its fugitive emission monitoring program, obliging operators to visit sites and scan for emissions with OGI cameras and fix methane leaks. As initially proposed, this was an expensive procedure and an Alt-FEMP approach has been developed that combines OGI cameras in specific areas, aerial detection with LIDAR, and ground-based, truck mounted equipment. Modeling techniques showed that a combination of different frequencies and resolutions gave better outcomes at lower costs. Canada Natural now uses the University of Calgary’s PoMELO technology, a truck-mounted multi sensor package that is used by operators to scan for emissions during regular site visits. Aerial Lidar from Bridger Photonics also ran. Canada Natural is also exploring the use of satellite technology to add another layer of high frequency/low resolution detection at ‘minimal cost’.
Next year’s PTN Events Oil and Gas Digital Transformation Conference will be held virtually on 12 - 13 September, 2023.
© Oil IT Journal - all rights reserved.