Keeping up with Digital Twins and Predictive Analytics
Digital Twins technology is becoming increasingly accessible to the oil and gas industry, and many refiners are keen to get on board with this digital trend. Asian Downstream Insights speaks to Anne-Marie Walters, Industry Marketing Director at Bentley Systems, to understand what organisations can do to effectively implement the technology, and the steps toward getting – and keeping – the workforce up to speed on their technological and AI skills.
To start off, please tell us a little more about Bentley Systems and what you do with regard to the oil and gas industry.
Bentley delivers software solutions that help owners, contractors, and operators of upstream and downstream assets reduce risk, lower costs, improve safety, and increase performance. Our solutions address the entire asset lifecycle, spanning design, construction, and operations for projects of any size and complexity.
What is the main driver influencing change across the future of the oil & gas industry?
The biggest driver in the industry is the drive toward net-zero carbon emissions, which is a huge challenge for all oil and gas companies and in which their future lies. Achieving net-zero operations requires a big shift across every aspect of an oil and gas company’s business. Digital technologies can help companies identify areas of existing operations through continuous monitoring, comparing situations to identify ways to reduce energy usage, improve reliability, and reduce emissions.
However, it will require changes to existing plants and the development of new services, such as transitioning from offshore oil and gas production to offshore wind farms and hydrogen production. Making these radical business changes requires the entire ecosystem working together to come up with innovative ideas.
What trends have been observed with regard to digitalisation in the oil and gas industry?
Digitalisation continues apace in the oil and gas industry, and achieving net-zero operations requires a big shift across every aspect of an oil and gas company’s business. As I said before, it requires the entire ecosystem working together to come up with innovative ideas. The oil and gas industry has always been a collaborative industry across its supply chain but still suffers from the lack of data sharing and interoperability, mainly due to closed systems used at various lifecycle stages.
An open approach to data sharing is required together with cross-industry data standards. I am seeing renewed efforts around cross-industry data standards, and we at Bentley are committed to delivering an open approach with our iTwin platform.
How can operators get started if they want to employ the digital twin concept?
For the oil and gas industry to achieve their goals for sustainability and net-zero targets, industry-wide collaboration is critical to enable better decision making and finding innovative solutions. Digital twins are the digital representation of physical assets, processes, or systems, and accurately reflect what the asset is doing, is designed to do, and how it has changed over time. Digital twins pull data from all sorts of systems. So, for the data in the digital twins to be trusted and dependable, the digital twin must accurately reflect reality.
Data is typically coming from different systems and silos, and digital twins enable that data to be aggregated and validated, so long as there is seamless integration. Bentley’s iTwin platform federates data, meaning that it exposes data in different systems into a single view, tracking changes and enabling everyone to share the same information. Open integration and interoperability through the digital twin platform is essential for everyone see the same data and collaborate.
The best place for a company to start, in regards deploying digital twins, is to focus on a specific business problem and gather the data required to solve that problem—start small and prove the value. However, keep the big picture in mind, as success will quickly be replicated. Without a solid foundation, rolling out digital twins (or any technology, for that matter) can become expensive and hard to maintain. Think big, start small, and then scale fast.
What metrics are available for operators to prove the ROI of implementing Digital Twin technology?
We are seeing a lot of companies focusing on standardisation efforts to ensure that they get good data from their contractors, using digital twins to verify and validate this data. Simply being able to pull data together from disparate sources—as well as verify and visualise the information easily with a good underlying data management processes to ensure data is updated as required—is giving them huge benefits. Projects are starting up faster, and delays caused by lack of information or slow approval processes are disappearing. Everyone has more confidence in the data, with some saving 20% 30% of their time in finding and verifying information.
An example is the engineering firm Hatch, which delivered a sulfuric acid plant in the Democratic Republic of the Congo three months ahead of schedule. They brought the plant to name plate capacity within one week of hot start up, something that would typically take six to 18 months. They achieved this result by using Bentley’s digital twin technology across the entire project lifecycle, from concept to commissioning, with 100% paperless delivery, which has now become standard practice for them.
Another operations user that comes to mind is Oman Gas, which is now part of the new OQ group. A few years ago, they implemented our AssetWise technology to address issues of reliability on their existing plants. They started on one plant and learned what critical data was required and how to best to address the cultural changes needed with their frontline maintenance workers. Today, they have successfully rolled out AssetWise Reliability across many of their plants and are achieving the reliability goals that they aimed for, as well as the associated cost savings in maintenance (around 9% cost savings). As a next step, they, together with the other groups within OQ, are looking to leverage digital data and technology to further improve plant performance, increase efficiency, and look at what can be done to meet climate goals, including reducing their methane and carbon dioxide emissions.
One of the biggest causes of both methane and carbon dioxide emissions in the oil and gas industry is flaring, which happens when a plant is shut down. The reliability program at Oman Gas has significantly reduced unexpected shutdowns and, consequently, reduced flaring. Moving forward, the group is looking at how to use digital twins to pull data together from all sources, as well as make more informed decisions about operations, maintenance, and capital projects.
How important is culture when implementing Digital Twins into the business infrastructure, and how can downstream business leaders ensure that the technology is accessible and understood across all parts of the business?
A big challenge when implementing digital twins is the move from a document-centric approach to information to a data-centric approach. Even though documents—such as drawings, spreadsheets, PDFs, and Word files—are often the preferred way of consuming information, the data relating to these files must be structured well and content updated so that everyone can find the relevant information easily and ensure that it’s accurate.
Keeping it simple is key, and everyone must take responsibility or correcting bad data and information. This will, however, happen naturally if most of the information is good and of value to the user. Focusing on the relevant data and information to solve a business problem is key, together with good data structuring for the long term via consistent data standards.
One of the biggest challenges the market faces is equipping the workforce to adapt to the pace of digitalisation. How can organisations empower the workforce to adapt to these strategies needed to survive and thrive in this new normal?
This is the perennial problem with digital technologies, which constantly outpace an organisation’s ability to successfully implement them. A proven way to gain acceptance and adoption is focusing on the needs of the front line, working to automate workflows that cut down on manual tasks. Together with trustworthy data and information, anything that makes life easier will empower the workforce. Many organisations are finding that Bentley’s iTwin platform—with its open, interoperable approach to easily pull data together from many disparate sources and flexible options for user interfacing—is enabling them to build innovative solutions that meets their specific worker and business needs. Some great examples are mentioned on our website.
How do you see predictive analytics developing in the industry’s near future?
The ability of the Bentley iTwin platform and solutions powered by iTwin, such as PlantSight, to pull data together from multiple sources opens up many opportunities for predictive analytics. In fact, PlantSight offers out-of-the-box analytics capabilities to transform data into actionable insights, supporting things like remote inspections, risk identification on projects, and emissions calculations.
A poll across the industry was conducted last year, and revealed that that 56% of senior AI professionals considered that a lack of additional and qualified AI workers was the biggest hurdle to be overcome in terms of AI implementation. How can operators manage to plug this talent gap?
Successful artificial intelligence (AI) depends on data—both quality and quantity—and business knowledge to verify the results. It does not mean that you need skilled AI professionals. By pairing IT professionals, who have good data skills, with experienced engineers or operators and using an intuitive digital twin solution, an organisation can develop and implement AI to support decision-making. As mentioned before, the key is to start small with a specific business problem and focus on the data needed. The IT person will know what is the reliable data while experience engineer or operator will know what is a good prediction or result.
How will predictive maintenance technologies evolve over the next three to five years? How will predictive maintenance help operators cut down on costs and unnecessary shutdown time?
While there are many that believe that AI and machine learning will automate many maintenance tasks it is actually the combination of technology with knowledge that will deliver results. The Internet of Things (IIoT), sensors, and digital technologies like cameras and image recognition can help gather the data while analytics helps sort through the masses of data, which sometimes can overwhelm operations and maintenance staff. Digital twins, coupled with captured experience (such as potential failure information), can help identify trends and performance differences, as well as help knowledgeable staff make better decisions.
How will IIoT impact risk mitigation and physical safety processes in plants and refineries?
As mentioned above, IIoT, sensors, and digital imaging technologies can gather plant data without a person being present. We are already seeing such technologies used with reality modelling to gather data about the condition of the existing plant, saving trips offshore or to the physical plant. This use case example from RocketMine is a perfect example of both improved safety and costs savings. Digital twins with reality models present can visually detail plant information to a remote engineer or inspector.
How can digitalisation/new technologies help organisations reach their sustainability goals?
In addition to saving trips to the plant – which, if offshore, significantly reduces organisation’s carbon footprint – open digital twins are readily accessible and enables teams to collaborate and innovate with new solutions to sustainability challenges. I believe that we will see many more innovative digital twin solutions coming to the industry, which will help organisations reduce methane and carbon dioxide emissions in real-time, as well as collaborate to develop new processes, such as using subsea hydrogen production to store marine energy.
How can AI innovations help the operators achieve their sustainability goals?
AI can help operators focus on the data and information that really makes a difference to their goals, as well as support their decision-making, rather than waiting for reports and results. However, AI needs knowledgeable workers to verify the models.