Plant reliability in the post-digital era

In this article:

• Increasing the stakes
• Will standards evolve, too?
• Holistic approach
• Future challenges
• Actionable advice

Digitalization is gaining momentum among manufacturers. Outdated processes are getting a much-needed overhaul. A new order of highly efficient, automated factories competing on a global scale is approaching. But what will a fully digital world mean to the plant maintenance industry? Will the factory of the future be self-running, self-diagnosing, and self-healing? Or will a limited number of skilled technicians struggle to keep the complex network of robotics and machinery working as needed? Predicting tomorrow’s response to escalating maintenance demands is no easy feat. But we know this: For digitalization to be effective, the plant maintenance strategy must also keep pace with technological change.

Forward-thinking factories understand the potential impact of technology on profitability and are embracing digitalization. Proof-of-concept projects have been completed, and full-scale deployments of new technologies are rolling out. The internet of things (IoT), embedded sensors, machine-to-machine connectivity, supply-chain networks, automated warehouses, and robotics are playing critical roles in driving shop-floor automation and improving efficiency. Even fundamental facility infrastructure elements such as lights and HVAC systems are becoming equipped with smart technology so that energy use can be optimized. Vehicles and material handling equipment, too, now have self-driving and self-monitoring capabilities.

Factories are hardly the labor-intensive, mass-production assembly lines that they once were. Escalating customer demands and heightened global competition have generated a “survival of the smartest” landscape. A need for creative problem-solving has led to expanded use of robotics and other automation-focused technologies, with these tools not necessarily replacing employees but often supplementing or complementing their work.  

• Read "The multiple dimensions of digital transformation"

The mainstream adoption of artificial intelligence (AI) and machine learning (ML) applications has spurred wider use of smart machinery. These “smart” assets, from temperature-controlled warehouse systems to visual monitoring tools meant to support quality control, can be connected to a central solution, given operational parameters, and allowed to self-manage assigned work cycles. If properly deployed, most machines need little supervision. This allows personnel to focus on more-creative and challenging tasks.

But, any machine, no matter how advanced, will require preventive maintenance, calibration, updates, safety inspections, and periodic repairs. Many pieces of equipment require parts such as bearings, belts or valves to be replaced based on wear. Others need consumables, like oil or filters, replaced routinely. And in high-tech machinery, even routine tasks such as checking fluid levels require expertise and specialized training. No manufacturer wants to send a new apprentice to work independently on a multimillion-dollar piece of equipment, even if it is only to check the hydraulic fluid. In digital factories, these demands will only increase, placing greater strain on the maintenance team.

Increasing the stakes

Innovation, particularly AI technology, is continually being refined, becoming easier to implement and easier to use. Templates for do-it-yourself AI applications help developers as well as front-line users create their own use cases. This democratizing of advanced technology will lead to more and more physical assets being connected to the network, sharing data, and consuming resources. More automated machinery means more assets to be monitored and maintained. Each asset is like a link in the systemwide chain. When one link breaks, the entire end-to-end process fails.  

Complex machinery, a hallmark of digital factories, requires highly trained maintenance crews. Failure of a critical piece of equipment can be disastrous for the digital enterprise, halting all work. Think of a high-temperature curing oven that is two stories tall and weighs tons. In a less-automated, old-school world, a wizened operator may have been able to use experience and instinct to manually compensate for temperature fluctuations or other idiosyncrasies that can arise in aging machinery. But in a highly automated digital factory, machinery is likely controlled by electronics and software, not levers and dials. If sensors detect a slight deviation from standards, an auto-shutdown sequence may be triggered, locking down the machinery until the issue is resolved and reinspected. Alternative manual processes may not be an option.

Will standards evolve, too?

Many maintenance teams strive to follow industry best practices, but most organizations still have room for improvement. Deloitte estimates that poor maintenance strategies can reduce a plant’s overall productive capacity by 5%–20%. Unplanned downtime costs industrial manufacturers an estimated $50 billion each year. 

In the maturity model for asset maintenance, a prescriptive approach is considered optimal. This will be even more important in the post-digital era. In this approach, advanced enterprise asset management (EAM) solutions suggest preventive tactics, prescribe how to act, and predict the outcome. Prescriptive maintenance uses predictive science and algorithms to provide a glimpse into the future and anticipate how the asset’s performance can be optimized.

As technology evolves in the next decade, maintenance solutions likely will evolve, too. This has happened in the past. PAS 55 was originally produced in 2004 by several organizations under the leadership of the Institute of Asset Management. It then underwent a substantial revision with 50 participating organizations from 15 industry sectors in 10 countries and evolved into PAS 55:2008. The PAS gives guidance and a 28-point requirements checklist of good practices in physical asset management.