Every day, businesses all over the world are generating millions of data points that reveal how their business functions and performs. Despite the growth and increasing availability of this data, businesses still struggle to connect their data, comprehend what their data is telling them and take real-world action as a result.
Digital twin technology is the missing link between this data and creating actionable business insights that can be used to improve business performance and outcomes.
So what exactly is digital twin technology and how does it work?
Louise Anderson, Executive Manager in Commonwealth Bank’s Emerging Technology team, explains that a digital twin is a data-driven digital replica of a business or a process. Instead of using traditional process mapping to understand how different features of a business interplay with each other, a digital twin allows the data to express that.
Once a digital twin has been created, Anderson says it’s possible to do any number of things such as: using it to simulate how changing decisions or processes within a business impact performance or introducing new variables and measuring the impact.
“Once you’ve started to understand the impact of making certain decisions we can then look at using digital twins to apply optimisation,” she says. “So it always starts with: build the model using available data, figure out how the different decision points within that are working and interacting with one another, identify the objective you want to hit within the business or process, and use [the digital twin] to figure out how to do that,” she says.
Digital twin technology in agriculture, manufacturing and financial services
Logistics companies have been using digital twins of their warehouses to look for inefficiencies, safety issues and wastage for several years. Similarly, supply chain operators use the technology to provide real-time monitoring which allows them to respond rapidly and add a fix to manufacturing lines and distribution flows.
In the agriculture sector, a digital twin collaboration between the CSIRO, Microsoft and Australian start-up Agronomeye is reshaping how Australian farmers are thinking about their future. It’s a 3D farm model, called AgTwin that brings CSIRO’s domain science together with digital integration and converts data into real-time insights for farmers. It’s easy to interpret and incredibly useful in decision making.
CSIRO’s data comes from its Boorowa Agricultural Research Station (BARS) north of Canberra.1 BARS uses digital agriculture technology for multiple lines of research, including conducting long-term farming system experiments that investigate moisture or carbon changes to soil over time. Agronomeye’s team creates a detailed digital map of a farm and this AgTwin platform serves as a “digital home” for the farm’s assets and operations. The high resolution data provides unprecedented layers of information to help farmers navigate the future in ways they had never thought possible.
In the United States, NVIDIA’s digital twin of BMW’s car assembly system is another example of digital twin technology in action. Global teams cooperate using different software packages and tools such as motion capture suits to design and plan the factory in real time 3D. This ability to operate in perfect simulation has revolutionised BMW’s planning processes and allows it to more efficiently reconfigure its assembly plants to accommodate new vehicle launches. It also optimises worker ergonomics and safety.2
In financial services, Anderson’s team has used digital twin technology to learn more about the bank’s home lending process – mapping the end-to-end process from application to decision. The process is complex as it involves several steps and teams, with a mix of manual and automated operations.
“What we’ve done is essentially collate a lot of data that is generated throughout the process and used it to model the end to end process using digital twin technology. It’s like a routing problem to try to figure out the best way to sequence every application through the funnel so that you can get best results for customers at any point in time,” explains Anderson.
Part of the complexity lies in the different datasets, teams and platforms that underpin the process. She says it’s about getting different datasets to talk to one another in a way they haven’t before. “They haven’t needed to stitch that whole end-to-end process together before … it’s where you find the real insights on where there are opportunities to improve.”
A digital twin proof of concept was run and delivered very promising results, says Richard Watts, General Manager, Home Lending Operations.
“There were multiple optimisation objectives that were investigated to determine the effect and the potential trade off on different metrics that we operate within the business. We are able to run different scenarios in a virtual environment and understand the impact of these changes before we make them,” Watts says. “In this way, Digital Twin technology can help de-risk some of the decision we make within the portfolio and help us uncover new ideas that we might not previously have come up with.”
The future of digital twin technology
Outside of operations, the technology has the potential to be applied to any organisational process and in particular those with increasing complexity and interdependencies.
This move is also shifting the conversation from predictive to prescriptive analytics. “It’s not just predicting what’s likely to happen, it’s telling you why that’s going to happen and more importantly what you should do about it,” she says.
“It’s getting a lot more granular. ‘How do I help people take more control of the outcome the business is trying to achieve and give them confidence in how they get there?’
“I think you need the technology that can help connect the dots and apply very complex modelling and mathematical scenarios to get those outcome and this is the opportunity Digital Twin Technology provides us.”
Our digital twin tech expert
Louise Anderson is Executive Manager, Emerging Technology and Open Innovation at CommBank. She leads a portfolio of innovation initiatives focused on accelerating the adoption of new and emerging technologies across the bank. Within her role, she primarily works with CommBank customer facing business units, taking an experimentation-led approach to designing and developing new products and digital experiences for customers. Before moving into emerging technology, Louise worked in strategy development with the CommBank Institutional Banking and Markets team and with Fidelity International in London.
