The arrival of the Fourth Industrial Revolution, as a solution to increasingly complex work environments, has heralded a new era in manufacturing. The introduction of cognitive and intelligent processes has opened up the possibly of a wide range of productivity enhancements which are driving the digital transformation of the manufacturing sector.
This has in turn brought up its own challenges, as manufacturing processes become ever more digital and automated. Companies need to discover how to leverage the potential of modern technological advancements such as the Internet of Things (IoT), cloud computing and machine learning to create both operational and strategic value and to gain an advantage over their competitors. One method that will be invaluable in facing these challenges, both in the present and the future, is the concept of the Digital Twin.
A Digital Twin is a near-real-time representation of the physical attributes of a factory, product, or process. Its application can be found in a wide range of fields. Through the connection of objects generating real-time data, a digital footprint of a product, from its design and development stages through to the end of its production cycle, can be developed. Data can be processed and aggregated based on a range of criteria and presented in various forms depending on specific use-cases. Over time, an evolving digital profile based on historical and current behaviour emerges. This can lead to the development of richer models and realistic measurements of a system’s inherent unpredictability.
In the past, it was not feasible, both from a technical as well as from an economic point of view, to handle the massive amounts of data required to generate highly realistic digital copies of an object. In recent years however, the costs of computing, storage and bandwidth have dropped dramatically. This enables the rapid collection, aggregation and analysis of data, generated by connected, smart technologies.
A well-constructed Digital Twin can facilitate the simulation of possible scenarios and predict possible outcomes, allowing one to try new strategies without impacting production. Not only does this save valuable time, but it can also help to identify possible flaws with a prospective strategy before it is deployed to the physical processes. By analysing the performance of machines and the quality of a product at each step of production, processes can be optimised to make better products at lower costs. The near-real-time tracking of components can alert relevant stakeholders to the eminence of breakdowns, allowing for possible issues to be averted before they actually happen through Predictive Maintenance. A Digital Twin can add further value through facilitating the remote performance of assistance and other service tasks.
Case studies of the digital twin
We are already seeing Digital Twins being applied in a wide variety of manufacturing industries. For example, in the automotive industry, Digital Twins can be used for a selection of car models, generations and configurations to model the number and type of microcontrollers, sensors and actuators as well as the installed software components. Using a digital twin, manufacturers can automate the process of data collection depending on the required platform to harmonise the various data models. This allows Data Scientists to access specific data points (such as vehicle speed) without having to know the technical details of the target platform.
In the realm of Connected Infrastructure, a ‘Device Shadow’ (a JSON document that is used to store and retrieve current state information for a device), can be used for devices for which a connection cannot always be guaranteed. The devices run on battery power and their state is managed in the cloud. In this way, the user is able to ensure synchronisation between the desired and actual state of the device, and to update the state of the device when needed.
Digital twins can also be used as a monitoring system for a fleet of assets in a factory. Thanks to individual signal sensors, alarm logs and complex state machines, manufacturers can now track a number of KPI’s; the most important of these being the overall equipment effectiveness (OEE). The OEE describes how well a machine is utilised relative to its capacity, and poor values on this metric signal that some kind of action such as machine maintenance or process optimisation is needed. Additionally, alarm logs can help users understand technical problems with individual machines and utilise alarm patterns to improve performance.
- Digital twins, AI-driven automation and blockchain will all change how companies see themselves in 2019
What role does the Digital Twin play in Industrie 4.0
Industrie 4.0 represents the next step in the evolution of manufacturing. We have gone from purely physical systems, through augmentations with digital systems followed by interactions between the physical and digital systems. Now we have the physical and digital systems converging; with a Digital Twin able to accurately represent a physical system.
In this context, Digital Twins can be used to evaluate production decisions prior to their implementation and to visualise performance. Machines in a factory can also receive updates to their software and configuration wirelessly, eliminating the need for some levels of on-site support, and allowing for a centralised team to perform service tasks at multiple sites simultaneously, even in remote locations. By connecting different systems and processes, a product can be tracked and monitored through the its entire production lifecycle. This opens the window to process optimisations, higher quality goods and reduced costs. With processes growing ever more complex, the Digital Twin offers a unique opportunity to gain control of an entire system.
Digital Twins can have the largest impact during the Production and Design phase; out in the field; and in the development of future products. The biggest advantages are through visualisation, which enhances human learning and decision-making processes and helps uncover areas which need attention; and collaboration, due to the fact that more stakeholders can gain an understanding of their machines and processes, no longer hindered by physical distance to the machines.
Digital Twins are the present and the future, with their prevalence beyond the world of manufacturing hinting at how commonplace they are likely to become within the industry and elsewhere. In our daily lives they are already having a profound impact almost without our knowing. Google maps, for example, is a digital twin many of us use on a daily basis; mapping the real onto the virtual in a way we engage with in a very human way. The same is happening in manufacturing as we have seen: an early adopter, perhaps, but by no means one that holds a monopoly on its application. Ultimately, Digital Twins are here to stay and with untold opportunities for those able to leverage this exciting technology.
Vincent Ohana, Partner, Concept Reply