Predictive Maintenance: NOT just a technological challenge | Article
PREDICTIVE MAINTENANCE: NOT JUST A TECHNOLOGICAL CHALLENGE
In the latest years the topic of predictive maintenance has gained a lot of interest among manufacturing companies from all over the world. The increased resonance of the subject is also confirmed by the fact that over the course of three years the monthly queries on predictive maintenance have become sensitively more frequent, as shown below.
Predictive maintenance nowadays enabled by machine learning and predictive analytics which represent one important way through which Internet of Things can create value in the factory setting.
According to Manyika, James, et al. “Better predictive maintenance can reduce equipment downtime by up to 50% and reduce equipment capital investment by 3 to 5%. A worksite operation can reduce maintenance costs by 5 to 10% and increase output by 3 to 5% by avoiding unplanned outages. In manufacturing, these savings have a potential economic impact of nearly $630 billion per year in 2025.” (“Unlocking the potential of the Internet of Things.” McKinsey & Company)
However, an interesting survey by PWC reports that only 11% of the surveyed factories have effectively deployed a PdM 4.0 approach, involving the application of big data analytics and machine learning techniques to predict failures that were unpredictable before the introduction of these IoT instruments.
The main barrier hindering a high maturity level PdM 4.0 strategy is the lack or improper use of the IoT data collected due to the absence of a digital culture. Several sources reveal that even if the industrial plants generate a big amount of data, quite often companies fail in identifying the necessary data to be collected, in storing, analyzing and using them for optimization and prediction purposes.
To put in place PdM 4.0, companies should first build up a predictive maintenance model by adopting reliability models (eg. RCA and FMEA)) and condition monitoring systems. Secondary, they should equip themselves with adequate Big Data and IoT infrastructures.
However, the implementation of PdM 4.0 is not just a technological challenge, as it encompasses also other aspects such as production strategy and management. Ultimately, predictive maintenance is a matter of constructing a digital culture within the company, which implies the employment of reliability engineers in predictive maintenance and data scientists, the creation of a cross-functional cooperative environment and the confidence with data-driven decision-making.
Therefore, PdM cannot be implemented only within the industrial plants, but it must be embedded into an overall company digital manufacturing strategy within a digital-minded environment.
References
Manyika, James, et al. “Unlocking the Potential of the Internet of Things.”McKinsey Global Institute (2015).
Haarman, Mark et al. “Predictive Maintenance 4.0 – Predict the unpredictable” PricewaterhouseCoopers (PWC) (2017)
Seebo, “Why Predictive Maintenance is Driving Industry 4.0” www.seebo.com
Related content