IoT Opens Doors for Preventive Maintenance, But Tricky to Implement


In manufacturing, one of the most pressing issues is how to address an equipment problem before it fails. The promise of the industrial Internet of Things is to address that very problem: Identify and fix a potential failure before it disrupts the entire system.

Preventive maintenance — and the parallel activity of predictive maintenance —  offer great potential for manufacturers but present a mix of complex challenges, even as industrial companies begin interconnecting their IoT systems to attain the hoped-for results of equipment efficiencies, stability and cost savings.

“Next-gen manufacturing equipment uses built-in sensors and sophisticated programming to perform predictive analytics and forecast potential issues before they happen,”Kayla Matthews writesOpens a new window  in the digital marketing outlet “Not only does this minimize downtime, but data-driven predictive analytics removes the guesswork from any preventative maintenance strategies. It also lets engineers schedule and initiate repairs when the machine is offline and dormant.”

The benefits of such predictive maintenance are significant: Poor upkeep can reduce a plant’s productivity by as much as 20%, according to a recent Deloitte & Touche report, which estimates that unplanned downtime costs industrial manufacturers about $50 billion each year.

Implementing a predictive maintenance strategy requires tremendous volumes of data to flow among teams, information technology experts and different business unitsOpens a new window within a company. A predictive maintenance strategy calls on companies to adopt new analytics technologies, either using their own expertise or adopting technologies that are preprogrammed with the necessary analytics.

>The data must be collected across a range of systems that include sensors, applications and infrastructure, using different communications protocols and a wide variety of data formatsOpens a new window .

“Organizations may also have solutions from multiple technology vendors rolled out on a site-by-site or project by project basis,” says Splunk Technology,Opens a new window  a data analytics company, in a white paper on preventive maintenance. “All of these factors result in data silos and swivel-chair integrations. The ability to leverage data across these silos in real-time is imperative to predict future conditions.”

Companies typically struggle as they move into the implementation stage of a preventive maintenance system with integration and technical issues. Successful preventive maintenance solutions need to be well-planned and tailored, with considerable thought and preparation dedicated to the particular needs of an organization.

“It should be understood that there is no ‘silver bullet’ computerized solution that will alleviate the need for some initial ground work, including determining exactly what type of maintenance organization and methodology makes sense for a particular operation,” says Chris Coleman,Opens a new window a manufacturing specialist at Deloitte Consulting.

Including the right tools, documents, people and appropriate skill sets are critical in ensuring that the technology solutions have the necessary support for the system to work. So, too, is a change management evaluation to make sure that the personnel have the necessary positive motivations to see the new system flourish.

A strong resistance to change may be prevalent in personnel who are highly trained and entrenched in operating procedures. A recent report Opens a new window by the Bain & Company consultancy recommends that vendors trying to work with companies on internet of things rollouts should focus on scalable solutions and  address specific roadblocks to adoption, including security, information technology and operational technology integration.

Another report by Amazon Web Services also suggests that vendors trying to sell a specific preventive maintenance solution might be better served by partnering with analytics companies that have built up a presence in cloud computingOpens a new window and artificial intelligence.

“Vendors shouldn’t assume their industry experience gives them a strong competitive edge in IoT,” Michael Schallehn writes in the Bain report. “They should partner with analytics firms and industry specialists to fill their gaps in knowledge and capabilities.”