Digital twins: Why smart industry is doubling up

26 Aug 2022

Image: © Lemonsoup14/

Prof Ed Curry explains why dynamic data models are key to future competitiveness.

The ability of top gun pilots to process information and make complex decisions faster than their opponent is key to achieving air superiority. Quickly reacting to a dynamic environment and making decisions with data has been a cornerstone of all pilot training.

US Air Force colonel John Boyd hypothesised that individuals and organisations undergo a continuous interaction cycle with their environment. Boyd developed the ‘OODA loop’ as a simple decision process by which an entity (either an individual or an organisation) reacts to an event by breaking the decision cycle down into four interrelated and overlapping processes through which one cycles continuously: observe, orient, decide, and act (OODA).

Digital twins are an emerging cyber-physical technology that can help entire organisations to make decisions like a Top Gun pilot by providing a holistic view of the world that integrates, correlates, and interprets data within the familiar human-friendly twin metaphor.

A digital twin is a digital representation that can be analysed to optimise the operation of the physical twin. It can be used to replicate physical assets (such as a car), processes (the value chain), a system (transport), or a physical environment (a building). It helps users observe the relevant data, orient to the best possible position relative to the goal, decide on the best course of action to take, and act rapidly on that decision.

Any system can be optimised by data-enabled simulation, and businesses are increasingly turning to digital twin technology to do just that.

‘We need to rethink how we process information and make decisions in dynamic environments’

Digital transformation in industry is currently driven by the convergence of two key fields: IoT and big data. Internet of things (IoT) technology allows for data to be captured via cameras, sensors, phones, appliances, street furniture etc in environments ranging from factories to shopping centres to homes. Big data analysis allows us to collate and derive knowledge from these multiple data sources in real time.

The ability of public and private sector organisations to effectively manage information and extract knowledge to make complex decisions is no longer a luxury. It is an imperative for survival and gaining competitive advantage. Many organisations are building their core business on their ability to collect and analyse information to extract business knowledge and insight. However, many are struggling in their digital transformation journey. If organisations are to maximise the benefits from the resulting data ecosystems, we need to rethink how we process information and make decisions in dynamic environments.

Digital transformation is creating the same problem for leaders in public and private organisations as Top Gun pilots face. They need to make better and faster decisions by processing large volumes of data, extracting insight, and then taking action using that insight. And it is not just leaders who face this challenge; the entire organisation must have this decision-making capacity.

Digital twins (also known as simulation models or data-driven models) are constructed from multiple data sources, including real-time IoT sensors, historical sensor data, traditional information systems, and input from human operators and domain experts. Feeding this data into a digital representation makes processes and operations visible for the purpose of analysis and optimisation.

Machine learning, cognitive analytics and AI techniques can then be used to learn the optimal operating conditions of the physical twin. These learnings can be applied to real-life operations in areas such as performance, maintenance, and user experience. In addition, digital twins are used to find root causes of potential anomalies (prediction) and to improve the physical process (innovation).

The digital representation updates and changes as the physical twin changes, continuing the process.

Digital twins can be built to represent anything from human organs such as the heart and lungs to aircraft engines and entire cities. However, creating digital twins requires organisations to have sophisticated capabilities in the areas of data management, artificial intelligence and data science.

Organisations need to bring their domain and data experts together to build meaningful digital twins that will enable data-driven innovation and creative thinking to drive their digital transformation. Siemens, GE, NASA and Singapore can’t be wrong.

By Prof Ed Curry

Ed Curry is a research leader at the Insight Centre for Data Analytics and a funded investigator at Lero, the Irish Software Research Centre. He is also vice-president of the Big Data Value Association, a non-profit industry-led organisation with the objective of increasing the competitiveness of European businesses with data-driven innovation.

Science Foundation Ireland’s Insight research centre helps organisations to create digital twins to support their digital transformation. Check out the website for more information.

10 things you need to know direct to your inbox every weekday. Sign up for the Daily Brief, Silicon Republic’s digest of essential sci-tech news.