What are Digital Twins?
Digital Twins are a digital replica of a physical asset or process and are used to simulate real-time or near real-time behaviour. This can range from the whole asset to just a much smaller scope size such as a single piece of equipment. Digital Twins are developed from a combination of IT (Information Technology) and OT (Operational Technology) and use real-time data from the physical asset to replicate and simulate real-life activity and their outcomes.
Through the collection of real time data via sensors and via data management solutions (e.g. data historians), a digital model can be developed and maintained to provide a virtual replica of the physical asset. As well the physical attributes (such as the dimensions), the behavioural attributes (e.g. performance, process conditions, and environmental factors) are all captured.
Digital Twins are frequently used for carrying out investigations and exploring what-if scenarios which can all be carried out in a risk-free environment and in a cost-effective way. Digital Twins are increasingly used in optimising and de-risking product development and supply chains. Some Digital Twins utilise predictive analytics and artificial intelligence (AI) capabilities to forecast future events, allowing for early identification of potential issues.
There are many value creating benefits of using Digital Twins that include:
- Personnel Training: Training in a risk-free, virtual environment.
- Performance Improvement: Monitoring and optimising processes.
- Faster Decision Making: Real-time data leads to quicker, more informed decisions.
- Defect Elimination: Identifying and solving issues without physical risks.
- Process Optimisation: Fine-tuning operations to maximise efficiency.
Some practical uses cases of Digital Twins
A Flight simulator is probably one of the most common examples. As part of a pilots’ training, a key part is to put theory into practice and so use a simulator to learn and to demonstrate the required competency and skill to be a pilot. A flight simulator enables a host of operational and emergency use cases to be recreated in a safe and cost-effective way. As part of the training, assessments can be carried out to review progress and be used for ongoing training and assessments.
Jet engines are another example of where having a digital replica allows the ability to monitor its operation and provide insights that will lead to increased reliability and reduce operational inefficiencies. These learnings will also provide value input into future design specifications and improvements.
Geographical Information System (GIS) is a geospatial Digital Twin that is used in many industries that include Oil & Gas, Automotive, and the Public Sector. GIS systems are digital replicas of a physical asset. In the Process Industries, any information (e.g. technical specifications) relating to the asset (e.g. a vessel or even a component such as a valve) can be linked to the asset. Any changes will be reflected so that the digital replica mirrors the physical version. Google Maps is one of the most widely used GIS platform. It is used for maps, directions, real-time information (e.g. traffic updates). Amongst its many functional use cases, we can use it to get a 3-D perspective of a location and identify physical buildings.
In the process industries such as Oil & Gas, Chemical and Manufacturing, Operations Training Simulators (OTS) are high value creation tools and are widely used as a result. Understanding the dynamics and operational behaviour end-to-end of an asset provides a wide range of benefits.
When training and assessing plant operators/technicians for example, real-life scenarios such as equipment failure, a vessel/pipeline blockage, or a power failure can be initiated as events in the OTS to see how the user would respond in a given situation.
For Greenfield (i.e. new) sites, an OTS provides an opportunity to validate Standard Operating Procedures (SOPs) and engineering studies to take place in addition to training activities. For Brownfield sites, they are useful for debottlenecking and feasibility studies. An OTS also provides an opportunity to execute operating strategies and identify defects.
From my personal experience of delivering Digital Twins, these solutions can have a large Return on Investment (ROI). Using the OTS as an example, operational issues can be resolved before they could possibly occur on the physical asset. In some instances where they were to occur, they would be very costly through loss of production and incurred operational downtime to resolve the issue and to bring the asset back online to normal production levels.
There are many more practical uses cases that can be discussed but the examples above provide a good starting point into some practical use cases and the benefit they bring in their respective application.
What are some of the challenges in developing, deploying and maintaining Digital Twins?
In the development phase, what can sometimes be challenging is getting the technical specifications of the physical asset required to build the virtual model. Close engagement with OEMs (Original Equipment Manufacturers) and support teams is very important. Also having a competent technical team skilled in both IT and OT domains to develop and validate and fine-tune the model through the build phase.
It’s always best to start small (e.g. a pilot study) from which learnings can be gained with reduced risk before scaling up. These early insights will help to avoid costly failures in larger-scale implementations. As an example, the pilot could be a very small part of an asset and the larger scale implementation being the whole asset.
A robust test plan needs to be developed to ensure all use cases have been thoroughly tested and validated. Taking the example of an OTS, a Factory Acceptance Testing (FAT) would be carried out after development. Site Acceptance Test (SAT) would be carried out after site installation.
A maintenance plan also needs to be developed for BAU (Business as Usual) activities to ensure the model is kept up to date with necessary software and security updates & patches.
Something that can easily be overlooked but is important is to a keep a Lesson Learned log to capture all learnings both in development and in operational use. These learnings can then be used in any future deployments and in Continuous Improvement (CI) initiatives.
Through incorporating learnings and continuously monitoring for improvement opportunities, Digital Twin implementations can become more robust, scalable, and beneficial for organisations.
Conclusion
It can be seen that Digital Twins are powerful transformative tools for improving efficiency, reducing risks, and facilitating innovation across many industries. By creating virtual replicas of physical systems, they allow for training, real-time monitoring, predictive maintenance and many other real-world use cases.As the technology continues to evolve, Digital Twins will have increasing value in their use and value creation.