Mayank Rausaria and Sandeep Kumar Sharma
Organisations continually strive to strike a balance between increasing equipment efficiency and reducing overall maintenance costs. Asset service engineers and quality departments across industries, such as airlines, oil & gas, pharmaceuticals, telecom, automobiles, and others, are pushed continuously to optimise their planning and operational expenses.
As the Industry 4.0 paradigm continues to evolve, the Digital Twins technology provides a pragmatic opportunity to leverage virtual models to also enable impact points such as to predict failures, prescribe actions, and support digital supply chain networks.
A digital twin, by definition, offers a seamless convergence of the physical and digital worlds — creating an ever-evolving digital profile of a physical asset based on its historical and current behaviour. For example: Supply Chain Optimisation through simulation and data monitoring.
The opportunities ahead
The idea of using digital models to optimise asset efficiency is not new. However, technological advancements in computational power, artificial intelligence (AI), augmented reality, actuators, and Internet of Things (IoT) are now allowing organisations to find increasingly sophisticated capabilities to simulate, model, and visualise intricate patterns.
The spectrum of possible instrumentation has also expanded beyond the conventional sensors and micro-controller to beacons, light detection & ranging, infrared, and vision technologies, allowing rich datasets to mine for signals. The next-generation technology has the potential to allow businesses to create increasingly sophisticated virtual models to optimise processes, products, and services.
It is imperative that organisations in the future would need to consider integrating IoT, machine learning, advanced computing infrastructure, and more to unlock entirely new business models, including as-a-service, pay-per-use, bundled services, and others.
Potential impact areas
Reachability, convenience, safety, appreciation for interdependent, and cascading, failures often constrain field maintenance. Digital twin technologies have the potential to not only enable remote sensing, the accurate curve of asset degradation, signal correlations that otherwise are hard to relate but also offer to design millions of complex what-if simulations and augment root cause analysis for detected real-world conditions.
Consider examples where an automotive original equipment manufacturer (OEM) is leveraging a state-of-the-art big data platform to predict failures and increase uptime of its asset’s aftermarket in real-time through an ensemble of machine learning algorithms. Similarly, a consumer product manufacturer is adopting a harmonised maintenance solution based on fault code correlation and image analytics for their welding and cutting robots.
Key bottlenecks in realising the promise of digital twins are interoperability of communication between sensors, operational technology hardware, and vendor platforms; suboptimal design of signal transmission; balancing the cost-versus-benefit analysis on the number of sensors needed. Organisations that plan to explore as-a-service revenue models should consider potential privacy and cyber risks as data get aggregated from multiple networks. Effective planning and architecture are seen to have yielded success in multiple enterprises.
The road ahead
Given the vast opportunity and applications of the digital twin, organisations often have a dilemma about where to start. While an overly simplistic model may not yield the value, undertaking a vast scope too fast can be overwhelming with millions of signals and complexity of technology.
Organisations could consider starting with a vision by exploring the possibilities for a digital twin. A focus on a specific use case that has a high possible value, and the best chance to succeed is to go broader than deeper, stabilise the pilot, and industrialise by adding adjacent process areas.
During the early phases of the journey, focus on consolidating and visualising complex data sets to generate rule-based alerts. Following this scale for anomaly detection, preventive maintenance, and prescribing corrective action capabilities can be looked into.
As capabilities and sophistication grow, it is expected to have more organisations use digital twin technologies to optimise processes, to make data-driven decisions in real-time, and to design new products and services. The potential gets amplified with advances in the near-future technologies such as 5G networks, open-source activity, and Quantum.
Globally, many organisations have already started pioneering and piloting digital twin technologies. Others are likely to soon follow as early adopters demonstrate the first-mover advantage in their respective sectors. The question for organisations should be on when and where to get started, rather than if.Mayank Rausaria is Partner, Deloitte Touche Tohmatsu India LLP and Sandeep Kumar Sharma is Managing Director, Deloitte in India. Views are personal.