IT in Manufacturing


Introduction to AI-based asset management

September 2025 IT in Manufacturing

For large, complex or hard to reach equipment, unplanned downtime quickly becomes a major problem. Predictive maintenance helps to minimise the probability of this occurring. The more detailed the knowledge about the current state of the plant, the more efficiently maintenance can be planned and performed. A digital twin creates the ideal conditions for this. The digital twin starter kit from Bosch and Pepperl+Fuchs offers an extremely simple introduction to this technology, and can also be easily used in existing plants.

If a geothermal pump needs to be hauled from a great depth, a wind turbine nacelle needs to be climbed, or a kilometre long conveyor belt system needs to be inspected for the repair, it will inevitably become expensive. The maintenance operation itself is very time consuming, and the production downtimes often contribute to additional costs. Predictive maintenance, which is intended to prevent or at least minimise downtimes, has so far often been linked to fixed cycles, and is initiated without precise knowledge of the actual plant condition. It is therefore not really predictive, but instead preventive, and tends to always take place earlier than necessary, or oherwise too late.

Modular solution for data-based prediction

Truly predictive or proactive maintenance requires insight into the plant and information from the field that is as precise as possible. In a digital twin, the available data is combined and can therefore serve as the basis for demand-oriented and efficient maintenance planning. In new plants, the digital twin is often implemented during installation. By contrast, existing plants feature a barrier for integrating additional technology into a functioning system.

The digital twin starter kit from Bosch Digital Twin Industries and Pepperl+Fuchs is specifically designed to overcome this threshold. It is a ready-to-use complete hardware and software solution that can be integrated into running plants according to the plug-and-play principle. The hardware consists of a powerful 12 kHz vibration sensor, an ICE2/3 IO-Link master, and the BTC22 embedded PC from Pepperl+Fuchs. Together, these devices establish a connection between the field level and a Bosch cloud.

Significant vibration plus field-cloud communication

The vibration sensors are included as standard to detect vibrations on machine and system parts. Their data can be used to draw far-reaching conclusions about the condition of individual components and larger units. They can be used to monitor critical wear parts in a targeted manner to determine any remaining runtimes. In addition to the vibration sensors, other sensors from the extensive portfolio offered by Pepperl+Fuchs can also be integrated to determine a wide range of process and status values. Device versions with approval for hazardous areas are also available.

The IO-Link master collects the sensor data from the field via an IO-Link connection and initially transfers it to the industrial BTC22 box thin client, which acts as an edge gateway. This fanless computer is not only powerful, but is also highly rugged and designed for continuous operation in harsh industrial environments. Using the pre-installed software from Bosch, it processes the sensor data from the field before forwarding it to the cloud in an aggregated manner.

Evaluation with ML and AI

The intelligent asset performance management tool is used to perform data analysis on the digital twin platform using a simulation model in real time. The platform is based on established cloud platforms such as AWS and MS-Azure. The analysis is performed using machine learning algorithms and artificial intelligence. Among other things, it enables the prediction of future failures, the optimisation of maintenance schedules, and the derivation of recommendation of actions. Mechanical problems can be predicted and visualised in detailed 3D models.

The kit is optimised for easy startup. The sensors, the IO-Link master and the BTC are installed according to the supplied wiring diagram and connected to the network. The initialisation process, in which the system automatically configures itself, starts automatically as soon as the voltage is switched on. After a short one-time programming phase, the process and device data is transferred to the Bosch platform, where it is processed and analysed. The operating state and further detailed plant data are displayed in the browser on the dashboard. Push notifications with information on critical situations can be sent to select recipients via email, SMS and other communication channels. The platform uses open communication standards and the system can be adapted to a wide variety of plant types and scaled as required.


Credit(s)



Share this article:
Share via emailShare via LinkedInPrint this page

Further reading:

Preparing Africa’s data centres for the demands of autonomous AI
Schneider Electric South Africa IT in Manufacturing
Africa’s data centre infrastructure must evolve significantly to support agentic AI workloads, with GPU-intensive computing, advanced cooling and reliable power demanding a fundamentally different approach to facility design and investment.

Read more...
Virtualising the control room to reshape building management systems
Schneider Electric South Africa IT in Manufacturing
Schneider Electric explains how virtualising building management systems frees facilities teams from ageing hardware, delivers stronger cybersecurity and enables portfolio-wide control from a single interface.

Read more...
AI infrastructure solutions for data campus
Schneider Electric South Africa IT in Manufacturing
Schneider Electric and Motivair deliver more than $290 million in power and cooling infrastructure for TeraWulf’s AI-ready Lake Mariner data campus.

Read more...
Africa’s data centre evolution: AI, edge computing and new energy demands
IT in Manufacturing
With less than half a gigawatt of active white space capacity serving over a billion people, Africa’s data centre sector faces a stark gap that AI’s surging power and cooling demands are set to widen. Vertiv and Open Africa Data Centres unpack how modularity, edge computing and smarter energy strategies could change the equation.

Read more...
How data and AI are unlocking South Africa’s next mineral frontier
IT in Manufacturing
Data, AI and blockchain are reshaping how mining companies explore, extract and prove the provenance of resources, opening a new frontier for South Africa’s mining sector.

Read more...
Luna Rossa partners with Siemens for 38th America’s Cup
Siemens South Africa IT in Manufacturing
Siemens details how Luna Rossa is using digital engineering and simulation tools, from CAD to structural optimisation, to design a faster yacht for the 38th America’s Cup.

Read more...
AI is raising South Africa’s cybersecurity stakes
IT in Manufacturing
As AI accelerates both the sophistication and scale of cyber threats, South African organisations face significant gaps in cloud security, data governance and identity management that need to be addressed before AI capabilities are further embedded into business operations.

Read more...
How modern EAM is becoming the driver of productivity, resilience and sustainability in mines
Schneider Electric South Africa IT in Manufacturing
Modern enterprise asset management is turning maintenance from a routine function into a strategic priority, helping mines extend the life of ageing infrastructure while improving productivity and safety.

Read more...
Physical AI solution with potential to transform South African manufacturing
IT in Manufacturing
NTT DATA and Hyster-Yale Materials Handling have deployed physical AI in an industrial assembly environment, embedding intelligence into production workflows to improve quality assurance and cut deployment timelines significantly.

Read more...
AI has an energy problem but Africa has an opportunity
IT in Manufacturing
Africa’s abundant renewable energy resources and largely unbuilt infrastructure put it in a unique position to power the growth of AI data centres sustainably, provided energy solutions are designed to match the scale and flexibility that AI demands.

Read more...









While every effort has been made to ensure the accuracy of the information contained herein, the publisher and its agents cannot be held responsible for any errors contained, or any loss incurred as a result. Articles published do not necessarily reflect the views of the publishers. The editor reserves the right to alter or cut copy. Articles submitted are deemed to have been cleared for publication. Advertisements and company contact details are published as provided by the advertiser. Technews Publishing (Pty) Ltd cannot be held responsible for the accuracy or veracity of supplied material.




© Technews Publishing (Pty) Ltd | All Rights Reserved