How predictive maintenance impacts the bottom line

The latest Siemens research suggests that Predictive Maintenance is already giving the bottom line of global manufacturers a major boost.

According to the ‘Readiness for Predictive Maintenance at Scale’ report, manufacturers running live projects are performing much better against a range of maintenance metrics.

The index below makes this clear. The green dots represent those firms at Level 5 of cultural maturity in maintenance, which includes live PdM.  

All key downtime measures are lower with PdM: 

• unplanned downtime incidents
• hour of downtime
• average cost per hour
• and annual downtime cost per facility. 

There are other benefits too. Less maintenance labor is required, and firms using PdM have a markedly better Overall Equipment Effectiveness score (OEE) than those with the lowest level of cultural maturity in maintenance. 

Predictive maintenance is arriving at scale

Historically, some manufacturers had concerns about the cost effectiveness of using PdM for every asset. This is changing fast.  

For example, one leading automotive manufacturer achieved a return on investment in three months by rolling out PdM at scale – without retrofitting a single new sensor. 

The manufacturer remotely monitors over 10,000 machines of 100 different types, using Siemens’ Senseye proprietary machine learning algorithms. These include robots, conveyors, drop lifters, pumps, motor fans, and press/stamping machines. 

Over 650 engineers and maintenance workers now use Senseye Predictive Maintenance to optimize maintenance activities and make repairs months before machine failure.

This manufacturer has:

• Made tens of millions of dollars in saved downtime
• Realized a rapid Return on Investment – less than 3 months 
• Up to 6 months advance warning of machine failure 
• Made year-on-year OEE improvements

This is just one project where the scale of the PdM deployment  goes well beyond monitoring critical assets. Today, PdM is being used at scale, with all classes of assets. 

Cost-effective predictive maintenance

Firms like the manufacturer above are realizing huge benefits from PdM at scale. This kind of success is now possible cost-effectively for two reasons. 

Firstly, in many cases there is no need to retrofit sensors. In many cases, predictive analytics can be run from existing data sources, such as Programmable Logic Controllers (PLCs). This dramatically cuts costs, partly of the cost of sensors and partly by cutting out the downtime associated with installation. 

And secondly, the need for expensive manual data analysis is being replaced by automated analysis based on machine learning algorithms, such as Siemens’ Senseye Predictive Maintenance.

Both developments are making the benefits of PdM at scale cost-effective. 

For more on how to implement PdM at scale, including the two key challenges and how to overcome them, read the full report here.