Future of predictive maintenance Part 3: Plant operations

The technological innovations for predictive maintenance covered in Part 2—from edge processing to advanced sensor capabilities—represent just one dimension of the transformation reshaping industrial operations. 

As the foundational technologies of predictive maintenance mature, new approaches to plant management and workforce development may extend far beyond traditional predictive maintenance.

In the present, however, many facility managers and maintenance teams face challenges in navigating the current, rapidly changing, technology.

This concluding section on the future of predictive maintenance looks at possible next-generation technologies such as digital twins and augmented reality. It also offers key takeaways and next steps for facilities implementing preventive maintenance.

This is the third installment of a three-part series exploring the future of predictive maintenance. Read Part 1 and Part 2. 

The future of plant operations

Several emerging technologies promise to reshape plant operations. However, their practical implementations will vary significantly based on industry needs and economic realities.

Digital twins: Promise versus practicality

Digital twins, or virtual replicas of physical processes that simulate real-world operations, are gaining traction in various industries. But, they often struggle with practical implementation at the plant floor level.

Digital twins can offer compelling theoretical benefits: 

• Modeling different scenarios 
• Optimizing capacity planning 
• Predicting equipment behavior without risking actual production

The technology would appeal to:

• Industries with consistent, predictable processes: Chemical processing plants, oil and gas facilities, and similar operations with steady-state conditions should benefit from digital twin modeling to understand failure modes and optimize throughput.
• Capacity planning and capital investment decisions: A digital twin could model different scenarios to determine optimal equipment configurations for major facility construction or expansion projects. 

For example:

A dairy processing facility could use digital twins to calculate:

• How many raw milk tanks would be needed for target throughput 
• If the facility can operate effectively with fewer processing lines 
• Equipment availability and performance levels required to potentially meet production goals with reduced capacity

However, digital twins may face significant limitations in more dynamic manufacturing environments. 

For example:

Assembly operations often involve constantly changing variables that would be difficult to accurately model. These include: 

• Temperature and humidity fluctuations 
• Unexpected maintenance issues 
• Human factors such as a tool left in or on a machine

Complex simulation modeling is still developing. For now, many organizations will often find more immediate value in addressing fundamental data collection and basic predictive maintenance.

Training the next generation with augmented reality

While digital twins address process optimization, augmented reality (AR) technology offers a more immediate solution for workforce challenges as experienced technicians retire and fewer young people enter hands-on manufacturing roles. 

Augmented reality, or AR, is technology that overlays digital information on a real-world environment. It can add sounds, images, or other sensory information to a physical environment, often through smartphones or specialized glasses. This immersive technology could provide:

• Step-by-step guidance
• Visual overlays 
• Real-time instruction directly in the work environment.

AR has shown success in consumer applications. 

For example: 

The Pokemon Go app, launched in 2016, still boasts more than 50 million users annually. It demonstrates that AR is intuitive, adoptable, and can provide contextual information to an environment.

Basic AR is already widespread in manufacturing.

For example:

Pick-to-light systems use illuminated displays to guide assembly workers. This simple form of augmented reality optimizes training and reduces errors in assembly processes. They prove that guided visual instruction can improve both efficiency and quality.

Similar AR technology in the maintenance world can:

• Direct maintenance technicians to the correct equipment components. 
• Highlight different components to guide a sequence of actions.
• Provide real-time feedback on procedure adherence.

For example:

AR systems would potentially identify bolts for removal, then highlight them one at a time so a technician removes them in the correct star pattern. The system might flash red warnings if someone attempts to skip steps in a critical sequence, such as removing the wrong bolt. It could also highlight the next component in green after each correctly-completed task. 

The most realistic path to adoption likely involves OEM collaboration with their customers. Equipment manufacturers could provide AR-compatible instruction files as standard deliverables with their machines.

Key takeaways and next steps

Predictive maintenance may take years to reach mass adoption. But, all trends and signs indicate that mass adoption of predictive maintenance is likely a question of “when,” not “if.” 

With the theories and technologies in place, the paradigm shift likely hinges on when those elements become more affordable and easy to implement. 

That means the organizations who begin implementing PdM now will be more competitive today and be positioned better to quickly adapt to technological improvements ahead of facilities that wait until predictive maintenance becomes widespread. 

Fortunately, the path to implementing a predictive maintenance strategy doesn’t have to be linear. Organizations with a foundation of condition monitoring and preventive maintenance are typically best positioned to make the leap. 

But, virtually any facility could take large steps toward PdM with the right strategy. A proof-of-concept project often focused on a few machines could create a track record of ROI and potentially empower a maintenance team to create new, scalable, processes. An established  ROI and workflow makes it easy to justify additional investments and  include more assets. 

Continue reading this series:

• Part 1: Mass adoption 
• Part 2: Advancing technologies