Digitalization is most apparent in the consumer market, where music downloads, online shopping and on-demand TV are hailed as marvels of the digital age. However, digitalization is also having a huge effect on manufacturing.

The ability to collect and manipulate vast amounts of digital information will catapult manufacturing into the future. By embracing digitalization, SKF is enhancing its core offering – bearings technology, and related services – so that its customers can further boost the performance of their rotating equipment. Furthermore, by focusing on industrial digitalization, the company aims to drive the further optimization of cost and efficiency of the full value chain, including World Class Manufacturing and Supply Chain integration.

Digitalization will affect all parts of the value chain, from design and manufacturing through to purchasing and maintenance.

SKF has been monitoring equipment remotely for around fifteen years and it now has around 1 million bearings connected to the Cloud. Data from them is gathered and interpreted daily, often with assistance from our experts. The ability to handle this data leads to enhanced analytics – allowing SKF to earlier detect potential failures in rotating equipment that affect overall equipment reliability and to get a better understanding of critical product and system design requirements.

The company has already developed platforms to help customers gather and interpret data. For instance, the Enlight platform helps operators visualize data from a variety of sources, using a device such as a smartphone or tablet. This is a smart way of putting ‘Big Data’ into an operator’s pocket.

The ‘connectivity’ of the data runs in all directions, and can be used in many ways. At its simplest, it connects a sensor to a remote diagnostics centre. However, the data – on the health of a bearing, for instance – can be fed right back to the design stage, and used to help redesign a better product.

Increased digitalization has also begun to allow more customized manufacturing. Because it can cut machine re-setting times close to zero, there are fewer restrictions to making customised products. Recently, the owner of an aluminium mill required bearings that would allow increased output – through a higher rolling speed – as well as lower maintenance costs and the elimination of unplanned downtime. SKF was able to produce four-row cylindrical roller bearings – complete with optimized surface properties and customised coatings – to boost service life and robustness, as well as designing out product cost.



Recently, SKF agreed a five-year ‘Rotation For Life’ contract with Zinkgruvan Mining of Sweden. SKF will carry out remote monitoring of four mills at a Zinkgruvan enrichment plant. The company will then pay SKF a fee – based on whether it meets its productivity targets.

This arrangement relies on digitalization technologies working in synchronisation. In one element of the contract, monitoring data from a conveyor belt is gathered automatically – with no human intervention – and an SKF specialist analyses the deviations if necessary, while a distributed lubrication system keeps the line running at optimum efficiency.

The ability to correlate a wider variety of data can further improve performance. For instance, the condition monitoring data that SKF routinely collects can now be combined with ‘process’ data such as machine speed and control parameters, through a collaboration with Honeywell. Combining these data streams has helped one of our joint customers – a major copper producer – to make more informed decisions on maintenance and asset performance.

The customer says that part failure would once have led to shutdown – but this can now be avoided thanks to the advance warning provided by the combination of process and monitoring data.

Having access to this wider array of data could enhance maintenance, and help customers to make more informed choices. For example, analysing both monitoring and process data might reveal that slowing a machine down by 3% would extend the maintenance period by four weeks. The customer can then balance a slight reduction in output with a longer production period – and make the best possible decision.

Automatic detection of a failing bearing is a massive step forward in efficiency. However, the process of ordering the replacement – including sending the purchase order through to manufacturing, estimating the lead time, and delivering the part – still involves major human intervention.

SKF is already gearing up for a future in which the faulty part effectively puts in an order for its own replacement. Because a smart sensor can already diagnose itself, it’s not hard to imagine that it might send an automated message all the way back through the supply chain.

It goes further than this: increased digitalization streamlines the manufacturing process. It has already helped to shrink machine re-setting times. In this way, a specific replacement part can be scheduled for addition to the production line with minimal disruption – and fast turnaround.

Combining these two factors – accurate prediction of a failing part, with ‘manufacturing to order’ – ensures that some ‘projected demand’ for parts is replaced by ‘actual demand’. This extends the ‘just in time’ manufacturing concept down as far as the individual component – and could one day bring stock levels close to zero. It’s hard to imagine a world without stock, but this vision is within sight.

This type of system is yet to be developed. However, SKF is running pilots in specific areas of the supply chain. In the future, the plan is to join these pilot projects together, allowing full, end-to-end digitalization.

The enormous power of existing digital technologies – such as smartphones – makes it easy to think that we have reached a pinnacle of performance. However, we are only at the start of digitalization within manufacturing. Every aspect of the manufacturing value chain can be enhanced by digitalization. Some have already emerged, while others are still on the horizon.

Can we really move from self-diagnosis of a bearing to self-ordering? Yes, we can: the hard part is predicting when it will happen.