The BMW Group has chosen to make use of sensors, data analysis and artificial intelligence (AI) to maintain its production facilities. Thus, rather than following the previous approach focusing on maintenance at regular intervals, predictive maintenance is carried out based on the current condition of the plant. This not only prevents unplanned downtime, but also positively affects sustainability and efficient use of resources, ensuring optimal system availability.

Cloud-based predictive maintenance solutions are currently being implemented across the group’s global manufacturing network.

By monitoring current equipment and parameters, predictive maintenance can predict failures before they occur. The data is used to optimize plant maintenance and determine when to replace components as a precautionary measure to avoid unnecessary outages. In addition, predictive maintenance improves efficiency and sustainability by ensuring that components that are still intact are not replaced prematurely.

Predictive maintenance uses a cloud platform to get early warnings about potential production outages. The data comes directly from the production plants themselves, which are connected to the cloud once via a portal, for monitoring and constant data transmission – which typically occurs once a second. The individual software modules within the platform can be flexibly activated and deactivated as required to adapt to constantly updated specifications. Thanks to the high degree of standardization between the individual components, the system is globally accessible, highly scalable and allows new application scenarios to be implemented and existing solutions to be quickly activated.

Predictive maintenance makes it possible to perform maintenance and repair tasks based on actual plant conditions within already planned production downtime. Repairs can be more accurate and efficient in terms of costs and resources.

See also  Illegal data collection: the CNIL sanctions Monsanto

Manutenzione predittiva: quando il macchinario rivela in anticipo le riparazioni necessarie<img src=” 1500w,×304.jpg 480w,×19.jpg 30w,×47.jpg 75w,×70.jpg 110w,×101.jpg 160w,×316.jpg 500w,×32.jpg 50w,×63.jpg 100w,×127.jpg 200w,×440.jpg 696w,×676.jpg 1068w,×420.jpg 664w” alt=”Predictive maintenance: when machinery reveals needed repairs in advance” width=”1500″ height=”949″ />

Flexible, highly automated mechanical transmission manufacturing plants produce one conventional motor or casing for an electric motor per minute. With the goal of keeping these machines in good condition, predictive maintenance uses simple statistical models – or predictive AI algorithms, in more complex cases – to detect any abnormalities. Visual warnings or signals are then issued to inform employees when maintenance is needed.

In body shop departments, welders perform about 15,000 welds per day. To prevent potential downtime, global welding machine data is collected by specially developed software. They are then transmitted to the cloud to be compared and analysed with the support of algorithms. All data is displayed on a dashboard to facilitate maintenance processes worldwide.

With regard to vehicle assembly, predictive maintenance facilitates the prevention of conveyor belt downtime. At the BMW plant in Regensburg, for example, the control units of the conveyor systems work 24/7 to transmit data on various areas – such as electricity, temperature and positions – to the cloud, where it will be constantly analysed. Data specialists can then check the location, condition and activities of each conveyor element at any time. Predictive AI models use the data to detect anomalies and pinpoint technical issues.

For all macitynet articles about cars, electric vehicles and smart mobility, see the Traveling Smart section of our website.

See also  Xiaomi Mi Pad 5 spotted live: here's what the back cover looks like