Digital acceleration: global manufacturing’s journey towards Predictive Maintenance

We recently published a free manufacturing and industrial sector report; The True Cost of Downtime, shares results from our global study on the impact of inefficient maintenance and unplanned downtime at many of the world’s leading industrial companies. 

Using data gathered from interviews with 72 engineering and IT professionals at multinational manufacturers, we have uncovered startling statistics on the actual cost of machine failure and insights into the appetite for Predictive Maintenance within these organizations.

We’ve also been able to assess their data readiness for generating a return on digital acceleration initiatives. 

Large-scale industrial facilities lose more than a day’s worth of production each month to machine failures costing on average $532,000 for every hour of unplanned downtime. Over a year, the average cost of this lost revenue, financial penalties, idle staff time, and restarting lines amounts to a staggering $172 million per plant annually.

Implementing Predictive Maintenance effectively here - using AI to analyze data relating to machine health and identifying early warning signs of deterioration - could cut downtime in half by directing the attention of maintenance engineers to where it is of most significant benefit.

The encouraging news is that almost three-quarters of the organizations we surveyed have least started on this journey, making Predictive Maintenance a strategic priority, and just over half (51%) perform some level of condition monitoring - an essential first step on the road to Predictive Maintenance. 87% have a factory historian collecting at least some of the data that support Predictive Maintenance.

Fully realizing the benefits of these initial investments, especially in condition monitoring, is not straightforward. It’s what McKinsey has dubbed ‘pilot purgatory’. Organizations can sometimes feel they’re stumbling forwards or even going backward because the initial investment still requires input from maintenance staff to get results. The enthusiasm here can quickly wane as fixing problems and other day-to-day activities start to receive more attention.

But at the same time, our study showed that only one in five major manufacturing organizations have a dedicated Predictive Maintenance team. And those already undertaking condition monitoring lost more hours (32 hours per plant per month) to unplanned downtime than those that didn’t (22).

This apparent contradiction is because many firms undertaking condition monitoring and moving towards Predictive Maintenance technology are driven to do so due to experiencing significant downtime problems. But it again highlights the fact that the road to generating a return on digital innovation is not always easy. 

Best practice on this journey is built on understanding how to get the relevant information, from areas that will show a benefit and then doing the hard cultural change work to realize benefits. It’s not about replacing engineering experts and maintenance teams but supporting good engineering practices with data and smart technology to deliver Predictive Maintenance sustainably and at scale.

Senseye’s white papers go into more detail on Predictive Maintenance best practices – and you can download and explore them here.


New call-to-action